This is hopefully the first post in a series of posts looking at the concept of “undiagnosing” patients. This might sound a bit daft, but it is something that we all do on a frequent basis – sometimes on our own patients, sometimes on others. Usually the original diagnosis is in error – not because of dodgy practice – but because we practice an inexact science, new information becomes available, and sometimes a fresh set of eyes can see the same data in a different light.
Undiagnosis can be very satisfying and challenging. Sometimes it requires a really thorough history and workup, sometimes just a rational look at the situation, sometimes you have to think a bit laterally and get creative.
I reckon I stop more oral antibiotics than I start in my ED practice, usually for simple URTIs – these are simple acts of “unprescribing” and unlabelling patients with benign problems who get exposed to unnecessary medical intervention and the potential associated side-effects.
My inspiration for this series has come from Dr Aaron Johnson. Aaron is a Canadian ED Physician who has ben a regular commentor on this blog. He now practices in the suburbs, but was a remote area GP / aeromedical doc on the Arctic Baffin Island. So a lot of similar challenges to what we see in Broome – but a slightly higher ice: pina colada ratio. Aaron also has a new blog that he runs for his trainees – check it out at Emergsource.com. It is just getting started, but from what I have seen – if you like Broome Docs – you will love Dr J’s blog.
If you have a case of Undiagnosis that you would like to share – send it through on the email email@example.com.
A 10 year old brought into the hospital ED after his second seizure. He had fallen to the floor and done some twitching and been unconscious for a minute or two, his parents had called EHS and he came to hospital by ambulance.
A week previously he was in the department with the same thing and had a normal CT head and been referred for an EEG and a neurology appointment. In the emerg I saw him in a non-cardiac-monitored bed that had been outfitted with seizure pads. He was now feeling fine and the parents were wondering if they could take him home and follow up with neurology as planned.
On detailed history of both events the boy had had a preceding feeling of anxiety and some mild chest pain. Both events were witnessed by the parents and they described him turning an ashen colour and falling to the floor, having possibly 30 seconds of shaking followed by 30 seconds of laying still. In both events the recovery was brisk with completely normal interaction and conversation within minutes of awakening.
The patient was on no medications, used no drugs and there was no notable family history. The patient had no other medical history whatsoever, and specifically no history of head injury, seizure, pre syncope, or exhertional symptoms of any sort. Physical examination was totally normal. I was concerned about this story on many levels; I was not happy that the supposed aura involved chest pain of any description, and I was not happy with the pre-collapse colour change. Most of all I was concerned with the seeming absence of a post ictal phase after such a seemingly serious event.
I transferred this patient to a cardiac bed and arranged investigations. His ECG showed an extremely long QTc of almost 600ms. His extended electrolyte panel was normal, his troponin was elevated at 0.25 and his lactate was 3.5. I felt he had likely had Torssades des Pointes with cardiovascular collapse as a result of congenital long QT syndrome.
He was a very hard sell to the cardiologists who felt he should receive a complete workup for seizure before they became involved. In the end and with much persuasion they saw him and put him on B-blockade medical therapy. This failed and he had another event and an implantable pacer defibrillator was placed with good result.
This was a great case for many reasons. It was in fact not a difficult diagnosis but was a good example of how sticky a diagnosis can be once attached. The cardiologists were initially convinced that this MUST be seizure, mainly on the basis of the fact that a seizure diagnosis had already been attached to the child. It was also a bit eye raising in my department when I wanted to do a cardiac work-up on the patient, and he had had a seizure, been placed in the stretcher we usually put seizure patients into and the bed had been outfitted with seizure pads. It’s made me much more cognizant of potential error associated with the problem of diagnostic momentum (and of course most importantly the patient was saved from early cardiac death).
Details altered a bit for HIPPA compliance. Dr. J
If you would like to hear more about this diagnostic challenge – Rob Orman has a good podcast on Paediatric Syncope at ERCast. This paper by Fischer et al, 2010 is a good review of the topic