I have a really special treat for you all. Drum roll….. A brand new guest contributor here on the Broome Docs blog!
Today’s case comes from one of my favourite FOAM folk – Dr Victoria Stephen. In case you have not met or followed Tori – stop what you are doing and jump onto her Twitter feed – @EMcardiac . She is my personal FOAM superhero – she works in one of the biggest, busiest and craziest hospitals in South Africa and loves to use the Ultrasound as much as I do! Incredibly humble & understated, Dr Stephen ‘walks the walk’ of which we hear so much talk about in the FOAM community.
So now – lets get onto the case. This is a trauma case. Imagine you are working in a really busy trauma ED in the bad part of Johannesburg. Stabbings, shootings and MVAs are rolling in all day. Your ED is stretched to the rafters with patients needing your care… in rolls the next patient.
A 33 year old pedestrian was brought to the ED by EMS. He has been hit by a car. The paramedics report that he is haemodynamically stable and alert with a suspected left femur fracture. A trac-3 splint is in place. A lodox scan is done in resus while he is still on the spine board and this is what it shows:
There is clearly a femur fracture on the left. How is his pelvis?
A minute later he is placed onto the resus stretcher. The patient is restless and complains of severe pain in his left thigh. He is rather pale and diaphoretic. Monitors are placed, recording a heart rate of 180-185 BPM, and the automated BP machine is having trouble recording his blood pressure. The patient does not appear as stable as previously thought. He feels cold to touch and his radial pulses are not palpable. A second large bore line is placed, fluids are started and blood is sent for crossmatch and blood gas analysis.
While the treatment is begun, a search for the cause of shock is underway. As ATLS teaches, the top three causes of shock in the trauma patient is haemorrhage, haemorrhage and…haemorrhage. So an eFAST is performed, and is negative. . There is no external blood loss from head wounds or anywhere else. The femur is already well splinted. A pelvic fracture is considered but the lodox scan has already shown that there are no pelvic or other extremity fractures. The mediastinum does not look wide and no pneumothorax is present. The blood gas comes back: pH of 7.32 pCO2 30 mmHG, Hb 12.0 g/dL and BE -10. After a litre and a half of crystalloid, his BP is 80/50, but the heart rate hasn’t shifted, racing away at 180-190 BPM.
A repeat eFAST is still negative. His BP is 100/65 now, but keeps dipping down to 80/50. Blood is being crossmatched, just in case, although a repeat blood gas looks much better. Despite this, the patient appears to be a transient responder to resuscitation. What could be going on?
Haemorrhage is still a possibility, particularly into the retroperitoneum. An eFAST is notoriously poor at making this diagnosis, and he’s just not stable enough for a panscan at the moment. The CT scanner is a fair distance away.
The patient may just be inadequately resuscitated. This may occur particularly in crush injuries and severe burns, if resuscitation is delayed or if the patient presents late. However this is unlikely in this case.
Another form of shock could be present. Obstructive forms of shock such as tension pneumothorax and cardiac tamponade have already been ruled out. Neurogenic is unlikely as he has no disability / evidence of thoracolumbar or neck trauma. This degree of tachycardia is pretty unusual for an adult. It also has not changed at all despite fluid resuscitation. One can’t make out the p waves on the monitor, either. A 12-lead ECG is done:
A myocardial contusion from blunt chest trauma is a good consideration. Dysrhythmias commonly occur as a result and most often manifest as sinus tachycardia, atrial fibrillation or ectopic beats. The patient denies any chest pain and no chest trauma is visible.
Thinking beyond trauma, recreational drugs and underlying illnesses should always be considered. He denies taking any recreational drugs; has no known medical conditions and denies having any palpitations. He was completely well until he hit the fender of that car.
Regardless of the cause or precipitant of the SVT, it requires treatment. Valsalva manoeuvres are attempted in the form of blowing a syringe and carotid sinus massage, but aren’t successful. Adenosine 6mg is given, which is followed by the seemingly eternal sinus pause. A sinus rhythm at 80 BPM commences thereafter and he remains in sinus. A 12 lead ECG done:
Casey: Recent blogpost on using CCB for SVT – this is clearly one situation where it would not be a great idea. Rare combination of events. This is where the protocols go out the window and you need to have a group huddle and think on the fly. Medicine – still an art, despite all the science!
The echocardiogram and CT angiogram of the chest are normal. Serial troponins are negative. He is transferred to orthopaedics after 24 hours of observation in high care and has an uneventful hospital course. He is referred to the cardiology clinic for further electrophysiological studies.
Casey: Now that is a really fascinating case. Dr Stephen and I had Twitter discussion about the medicine and the cognitive errors that cases like this bring to the fore.
So, let’s perform a further cognitive autopsy:
System 1 thinking is the rapid, intuitive form of thinking that relies on pattern recognition. It often serves us well, particularly in trauma resuscitation, where diagnosis and management often need to occur simultaneously. System 2 thinking which is slow, deliberate and analytical reasoning may not appear to have a place in the trauma resus bay, but is proven to be more accurate than system 1 when dealing with more obscure presentations. Surprisingly, even in a trauma resus, we have more time to think than we perceive. Particularly in unusual presentations, a cognitive checkpoint should take place where the team pauses to consider whether anything has been missed or not been considered.
Pattern recognition can be useful in leading to a rapid diagnosis but it can lead to anchoring bias, where an initial diagnosis is focused on at the expense of other diagnoses not being considered. In this case, where shock was recognised, it made perfect sense to begin treating and investigating for the most common cause of shock in trauma – haemorrhage. But the heart rate was too high and out of keeping of other clinical findings, such as the patient’s improving blood gas and perfusion. Another form of shock was present. System 1 thinking will prompt one to recall a familiar, common diagnosis first. In a dedicated trauma unit, an SVT was a foreign entity, so was not initially considered.
How can we improve our thinking in trauma resuscitation and reduce cognitive error?
- System 1 and 2 thinking both require a solid knowledge base to avoid error. Courses with intense training in trauma are a good start. There is no substitute for knowledge, training and experience.
- Be aware of cognitive biases in the trauma setting. Search satisficing, which is stopping to look for other injuries or conditions after one particular injury is found, is common. Examples include failing to look for more than one fracture after the first fracture is found or failing to recognise the potential of occult bleeding in the patient with neurogenic shock from a devastating spinal cord injury.
- Situational awareness as applied to the medical field is described as “the detection and interpretation of situational clues from multiple, dynamically changing sources (including from patients, other members of the team, and from displays and equipment), thinking about what might happen in the future and adapting to evolving situations”. Situational awareness is improved by the use of a Trauma Team leader to guide the overall resuscitation while other members are focused on critical tasks. Other strategies to promote situational awareness in resus teams to reduce medical error such as simulation, shared mental models are well described in this paper: “Beyond crisis resource management: new frontiers in human factors training for acute care medicine” by Andrew Petrosoniak and Chris Hicks.
- Performing a tertiary survey once the patient has been stabilised and life threatening injuries have been managed. This includes a head-to-toe examination looking for potential missed injuries such as a scaphoid fracture. It also may include further enquiry into the patient’s past medical history and consultation with other specialists as appropriate.
Patients often fail to read the medical textbook. It’s up to us to recognise unusual presentations, adapt our thinking and improve patient safety.
Interested to read / hear more on the cognitive side of what we do? Check out these links as a starter.
Ed: Of course if you like your cognitive psychology delivered in the theme of Star Trek by a cheeky, dimpled Intensivist then you should go back and see this talk from the original SMACC conference in Sydney 2013… “All Doctors are Jackasses” by Chris Nickson.
Thanks to Tori for sharing this case and the cognitive autopsy. These are exactly the types of cases where things can go wrong despite all the right moves. Sometimes we just need to think better – this is hard to do in the moment, but awareness and the discipline to stop and rethink are key in my opinion.
If you would like to hear more from my favourite African Doc then please let me know… I reckon we can twist her arm to do a few more?