SimFridays - pulmonary embolism

For October the sim theme is “breathing”. This blog covers some of the learning points from 29/10/20. We will be aiming to run simulations weekly - mostly Fridays but not always - see the gmail calendar. November will be "cardiovascular" month.
​

​The simulated case: 
Sam is a woman in her seventies presenting with increasing shortness of breath over the last 2 days. She is requiring >10L/min of oxygen by facemask to keep saturations >94%. 

At this point consider how wide the potential causes of breathlessness are. After treating the hypoxia, which tests or investigations might increase or decrease the likelihood of it being any of these potential diagnoses?

What happened?
We ran this short simulation with a nurse, HCA and trainee ACP in the resus area of ED. A history was taken, observations recorded and appropriate oxygen delivered. A range of causes were considered and appropriate investigations (bloods, ECG, chest x-ray) carried out. 

This simulated patient ultimately would have been found to have bilateral pulmonary emboli (they had increased risk due to metastatic cancer); however, in this short sim the intention was for the patient to be assessed, have emergency management and the right tests thought about. 

What did we think? 
In debrief we discussed:

History taking:
When asking rapid questions to narrow down the differential diagnosis, there is a risk with asking questions in the negative (e.g. “you don’t have chest pain at all?”) that the patient may passively reply “no” to questions, compared to “are you in any pain?” they may be more likely to explain that they do.

Investigating and treating PE:
We discussed the PERC score, Well’s score, ECG signs, and the treatment options for PE including anticoagulation, thrombolysis and interventional radiology - see guidelines section below. 

In terms of ECG signs, the most reliable is sinus tachycardia, however this article and its links cover well the signs of right heart strain to look for, and how to differentiate it from similar presentations. 

Decision making in ED:
Breathlessness (or even hypoxia) has a wide list of potential causes. In the emergency department patients are being seen often at the early stages of illness where the disease is potentially less manifest and information is scarce. At this point there is a much higher uncertainty. Prof Carley has a recorded talk and a blog about making decisions amongst this uncertainty here. 

In cases I have seen during this period of high uncertainty it may be that the patient is treated for several potential causes of their symptoms. For example the patient with PE and secondary heart strain may have already had antibiotics for ?sepsis and dual antiplatelets for ?ACS. This can be okay, as long as the decisions were made with good intentions based on the information available at the time. In the case of those treatments, potentially the benefit from early treatment and the high risk of not treating them may outweigh the risk of giving treatment to someone who is later found not to have the disease. 

However there are other risks lying in this period of uncertainty. We discussed in the debrief the potential for anchoring bias, where the clinician “anchors” to one early piece of information and all subsequent information is either thought to fit that mental model or is discarded. This may mean that the patient with PE is actually only ever treated for pneumonia, and PE is never considered. Personally, I suspect this bias has greater power when a clinical handover happens - if you are handed over a patient “we’re treating them for X, and they’ve been referred to MAU” is there a risk you anchor to that diagnosis? If new information comes along (e.g. new blood results, their chest x-ray, a colleague reporting they don’t seem to be improving) it’s important to go back and reassess with an open mind. Anecdotally someone I know who suffered a PE explained to me their experience in ED felt like minds had been made up immediately and subsequent information didn’t seem to adjust that idea. 

Another similar bias we discussed was confirmation bias: believing the patient has a particular diagnosis and then unconsciously only retaining information that supports this, discounting that which refutes it. A technique to combat these biases is to actively seek out information which would change your diagnosis or plan. What other diagnoses would be really important not to miss, and what signs might lead to that diagnosis instead? ​

In our new layout of ED, with front-door senior assessments, patients often have a potential diagnostic label attached to them before they reach the more junior clinicians. This has clear logical benefits for patients. But we raised in debrief that there is the potential for the biases above to occur following this. More junior members of the team may feel difficulty in broaching alternative diagnoses. So we discussed in general how one might explore decision making with a colleague by framing it as a “teaching moment”. For example, “I wouldn’t have thought about X diagnosis for this person, do you mind helping me understand why it is X and not potentially Y?” or “I noticed X piece of information, from my lectures they used to say X was associated with Y, but here you’ve said it’s most likely Z - do you mind telling me about why it’s different here?”. We’ve talked before in this blog about graded assertiveness and the PACE model here.

For a gateway into the larger field of ‘thinking about how we think’ in emergency medicine, go to this blog by Dr Natalie May. And for a deep-dive, I recommend this ebook on how we make decisions in the ED. 

The guidelines:
The EDIS guideline can be found under “adult medicine”, with other helpful resources being the British Thoracic Society guideline (note from 2003) and this LifeInTheFastLane article. These three have been used in the following sections. 

The PERC rule is a rule-out scoring system for low risk emergency department patients. A score of zero in a low risk patient means <2% risk of PE, which means the risks of investigating most of these patients further would outweigh benefits averaged over the population. It was not possible to use it in this case as it is only for use when the risk of PE is low (e.g. Wells <1).

A Wells score is a very important step in the investigation of potential PE as it helps us determine how likely the diagnosis is as a baseline before any investigation (the pre-test probability). We then seek to use examination and tests to change this probability up or down. A patient’s Wells score helps us decide whether a d-dimer blood test will aid us in the diagnosis or not. Because of the test characteristics of d-dimer, where pre-test probability is low a negative d-dimer can help rule out PE, but where the pre-test probability is high a positive or negative d-dimer will not significantly alter the probability of it being PE. Please do look at our EDIS guideline which has a flowchart on when to use PERC, d-dimer and imaging.

The patient in this scenario would have gone on to have a CT pulmonary angiogram. It’s worth noting that patients usually need a green (18 gauge) cannula for this. 

With the diagnosis confirmed there are different possible treatments. Thrombolysis is generally used when there is ‘massive PE’ i.e. with circulatory compromise, or in PE-associated cardiac arrest where a bolus of 50mg alteplase can be used. The patient in this scenario had normal blood pressure and had significant bleeding risks, so thrombolysis was not being considered initially. 

Interventional radiology can be used to remove clots. If sub-massive (inc heart strain) or massive PE has been detected, discuss with the ED senior or IR directly whether the patient is suitable. 

The EDIS guideline gives DOAC dosing or weight-based doses of enoxaparin if anticoagulation is being used. There is a separate guideline on the “outpatient pathway” that shows where someone can be safely discharged with treatment vs when admission is more appropriate.


To do:
Look at the EDIS PE guideline and the separate link for who can be treated with “the outpatient pathway”   [  ]

When looking after a patient in the next week try to think specifically about possible diagnostic biases and how you might acknowledge and avoid them based on the above  [  ]

If you took part in the sim, you can use this blog as a starter to reflect on your own experience of it   [  ]

​
James Keitley - ED Sim Fellow

---------------
For clinical decisions please refer directly to the guidance. This blog may not be updated. All images copyright- and attribution-free in the public domain.