Chest Pain

Chest pain is the bread and butter of Emergency Medicine! You will see this presentation every day. It’s important to have a structured approach so that you don’t miss any important details or get stuck in the weeds. 

Our role in emergency is to exclude life threatening causes and, of course, to act if we find one. Our job is not to find a reason for someone’s chest pain. That is usually not possible in emergency. Giving someone a diagnosis to explain away their symptoms can be dangerous. I will come back to this later.

Some would argue that you should always have a differential in your head, right from the triage note, and that every question you ask and every examination finding you look for should be getting you one step closer to, or away from, a diagnosis.

So, what should we be thinking about for chest pain? The list is very long and you will be able to think of many more differentials than what I talk about here. But for every chest pain patient I want to make sure you’ve thought of these:

• Acute coronary syndrome (unstable angina, crescendo angina, NSTEMI, STEMI)

• Pulmonary embolism

• Pneumothorax

• Aortic dissection

Pneumonia can cause chest pain but will look a little different – usually cough, shortness of breath and fever. And pericarditis is always something to look out for too but is something I don’t really want to cover in this blog (though check out this ECG article if you’re interested). The short list above is important because the diagnoses are life threatening and sometimes tricky to pin down.

Let’s start with Jake. He is a 29 year-old who didn’t really want to come to emergency but came because he’s had mild right sided chest pain all day and his girlfriend was worried. He thinks he’s fine. What do we need to do to make sure he is?

Then we meet Anja. She’s a 55 year-old woman with hypertension, a heavy smoking history and is reluctant to see GPs for anything more than a medical certificate. And yet she’s here in emergency today. What’s got her so frightened?

Acute Coronary Syndrome

The history is key. Besides a detailed history about the event that brought them to ED today, you should always ask about chest pain on exertion (angina), prior episodes of chest pain before today and cardiac risk factors. These are so important that I put them in my main history instead of past medical. They are almost the first thing I ask every patient with chest pain:

• Prior history of ischaemic heart disease

• Smoking history

• Hypertension

• High cholesterol

• Diabetes

• Obesity (I do not ask but rather infer this)

• Family history of cardiac events under the age of 65 (parents especially)

• Some ethnicities (given Pacific islander patients, for example, have a higher risk of ischaemic heart disease)

Someone like Jake probably has no history of angina and none of these risk factors. Anja has several. It’s also worth noting that patients who don’t see a doctor raise alarm bells for me. They’re inevitably the patients that turn up for help far too late and who won’t seek help if you send them home and things get worse.

It’s also worth noting that women with cardiac ischaemia frequently present with atypical chest pain, some studies quote up to 30% or even more. Does your female patient have jaw pain? Shoulder pain? Epigastric pain? Nausea and sweats but no pain? Please think about a heart attack.

Chest pain patients will get two ECGs at least. Why two? Any changes between the ECGs we see are called “dynamic changes” and are a clue that active ischaemia is happening, with or without patient symptoms.

And why two troponins? If someone is having a very acute ischaemic episode the first troponin is usually normal. The second troponin often isn’t. Any troponin rise is cause for consideration. It may not mean they are having a heart attack but there’s always a reason why it’s elevated.

When you have all of this information you can make a good judgement call about whether we ischaemic heart disease is probable, possible or unlikely.

The HEART score is very helpful for this. It predicts the risk of a big heart attack in the next 30 days based on the information you gather. I find it most helpful if it reads low-risk (these patients can be discharged with reassurance) or high-risk (these patients should always be discussed with Cardiology) but not so great when it’s in the middle. This score is really designed for low-risk chest pain. If your patient has had heart attacks before or has dynamic changes don’t even bother with this.

In addition to everything above, a chest xray is something I routinely order for nearly all chest pain patients. Young patients, especially those who have never had chest pain before, may have a spontaneous pneumothorax which you can’t diagnose any other way (besides from bedside ultrasound but that’s getting a bit too complicated). It can also give you lots of good information for comorbid patients including pulmonary congestion, cardiac silhouette and if they have pneumonia.

If you want to find out why we do serial troponins or any gritty details about ACS work-up in ED I can’t recommend EM Cases more highly. This episode is exhaustively detailed.

Pulmonary Embolism

In a sentence, if you don’t think about a PE you will miss the diagnosis. I ask about PE risk factors in every chest pain patient just to force myself to entertain the possibility. The Wells' score and PERC rule for younger patients are excellent clinical tools to help your questioning and you can put these into your documentation to show you’ve thought hard about whether this person has a PE or not. 

Some words of warning here, to use the PERC tool effectively you must believe the patient is very unlikely to have a PE. And in both tools, if the patient has had a heart rate over 100 at any time, then they score a point, even if it’s not persistent.

In my experience, patients with large PEs often have significant progressive exertional dyspnoea. If a healthy person is now getting puffed out when walking the dog, then that is bad news. I think it’s also worth noting that if someone needs supplemental oxygen, and you don’t have a good reason for this, then that person could have a PE.

I find D Dimers frustrating, usually useless, and I rarely do them. I would certainly discuss your patient with a senior before deciding to send one because they can muddy the waters. If you think it’s unlikely the patient has a PE, then just exclude this clinically. If you’re worried they do have a PE, then just do a CTPA. D Dimers should be reserved for patients where you need to exclude a PE but you’re trying to avoid a scan. Unfortunately, in tertiary centres most of these patients end up getting a scan anyway.

Pregnancy is an exceptional case. These patients may be managed best by senior doctors. PE is the leading cause of death in pregnant women in the developed world and it’s exceptionally difficult to diagnose clinically. Pregnancy is also a prothrombotic state, a risk factor for PE. The normal physiological changes that women experience as part of pregnancy (worsening shortness of breath, peripheral oedema) are signs that could signal a PE. We also want to avoid radiation exposure in pregnant women. It’s a minefield. This ACI article gives an excellent overview of the considerations and difficulties in this case.

Pneumothorax

Not much to say here except spontaneous pneumothorax occurs in the young and it’s hard to predict. Sometimes on examination you can feel subcutaneous emphysema in the patient’s neck (it feels a little like cellophane). These patients can look extremely well. It’s also much more difficult than you would expect to see this on an x-ray because they appear as just a sliver in some. When you look at a chest x-ray you have to ask every time “is there a pneumothorax?”.

Aortic Dissection

We classically think of aortic dissection as a ripping chest pain that radiates to the back between the shoulder blades but in all honesty, this is a hard diagnosis to make that’s also rare and life-threatening. EM Cases again has an excellent collection of pearls on this topic.

Bilateral blood pressures are not specific nor sensitive enough. 19% of the normal population have a BP difference of over 20 mmHg between their upper limbs. Migrating pain has a high positive likelihood ratio (i.e. makes the diagnosis much more likely). The concept of chest pain plus is also very helpful – patient’s may have another critical symptom with their chest pain like a focal neurological deficit or limb ischaemia to herald a dissection. At this point though, it’s likely your boss is standing right next to you.

Discharging Your Patient

So, you’ve worked up the patient diligently with a thorough history and exam, normal ECGs, negative troponin tests, a normal chest x-ray, a low Well’s score and they’re going home. “What do I have doc?”

Some doctors feel the need to give patients an explanation for their symptoms. It’s easy to say it could be reflux or maybe costochondritis. You’ll also never be able to prove it, and these assurances are not without danger. Patients have been known to hold on to these diagnoses and delay their return to ED, even if their symptoms change, because they were told it was nothing serious. It comes up in Morbidity and Mortality meetings more than you would imagine.

I tell my patients that I only get to meet them once and that if things change, we want to see them again. I tell them that I could always be wrong, something some doctors feel reluctant to say but that I find very powerful – patients appreciate the honesty. This safety netting is essential in emergency because of the uncertainty that’s baked into our work.