Headache
Posted: August 27, 2011 Filed under: Podcasts 4 Comments »We see patients with headache all the time in the ED. Most patients with headache don’t have a life threatening diagnosis but its our job to pick up that small percentage of patients that do. In this episode we’ll go through how to take a good headache history, how to catch the red flags, the workup, and treatment of headache in the ED. There’s also an extended bonus section that will review how to do an LP along with a few tricks of the trade.
Headache show notes (Word format)
Headache show notes (PDF format)
Carroll -
I had two things…
1. Read this article and tell me what you think. I consider this to be a practice changing paper; I am curious to see if this will be changing your future practice.
http://www.bmj.com/content/343/bmj.d4277
2. I really think that the terms hypertensive urgency/emergency need to be done away with. The term hypertensive urgency has absolutely no literature to support it; the term “uncontrolled hypertension” is a better term. No literature has found a need to address any blood pressure in an asymptomatic patient in the ED – it can be managed by their PCP. I would also argue that the term hypertensive emergency probably needs to be done away with as well. Almost every diagnosis that is associated with a hypertensive emergency (i.e. SAH, dissection, hemorrhagic stroke, CHF, etc.) is subject to a chicken and egg argument. There is no literature to support the notion that a headache + uncontrolled hypertension = hypertensive emergency; without anything to support the diagnosis of a bleed, etc. The only exception to this I can think of would be hypertensive encephalopathy and that is really a diagnosis of exclusion. What are your thoughts on this?
Schauer
Hey Steve
1) Funny you bring up that article- I just read it yesterday after seeing it laying around in the team center. I think that is 95% of the way there to being the perfect study. If they had done LPs on everyone instead of 50% then I would say it would be practice changing but since they did I have trouble accepting it. Yes, they claimed to have followed up 95%+ of patients in some sort of fashion to make sure that they didn’t have a SAH but the number was actually less than that- somewhere around 500 out of 2000 patients were not able to be directly contacted for some reason at 6 months. So they did the next best thing and tracked down as many medical records, death certificates, medical records of neurosurgical referral centers, etc. While this is a reasonable approach, it doesn’t prove beyond the shadow of a doubt that those people didn’t have an SAH. Based on the overall incidence of disease (7%) you can roughly expect approximately 35 patients in this 500 to have an SAH. I know that’s not the way that the stats work but this argument isn’t unreasonable. I thought there numbers were great (3000+) and it took 9 years- I really just wished that they had just done the LPs that they needed to do.
Also- keep in mind that there were SAHs detected at 8 hours by CT. Is that 2 hour buffer enough to hang your hat on? Does everyone know exactly when each headache started? Since you can argue that this study matched real practice you can say that this is irrelevant but having patient’s nail down the minute their headache started to avoid an LP is shaky practice.
The thing I object to in this article is in the “clinical implications” section they were suggesting that headaches need to be treated as “brain attacks” and we need to “change the paradigm” of how we deal with headaches….oh brother…so now it becomes a super emergency if you’ve had a headache for 5 hours- better get your CT scan now! Or get to the ER as soon as that headache starts- it could be a bleed! Don’t get me started…
2) I agree with you- hypertensive emergency, urgency, acutely elevated blood pressure or whatever we are calling it this month is a vague term that’s why I didn’t dwell on it too much. I agree- to say something is an hypertensive emergency, you need an actual diagnosis with a high blood pressure. I think we’ll see the term hypertensive urgency done away with although I think hypertensive emergency can be a good term because it helps you realize that a high BP with certain diagnoses needs to be lowered in the acute setting (whether this is quickly in a dissection or slowly in a stroke, its quicker than the days to weeks for asymptomatic hypertension). I just wanted to let people to know the terms out there and its a good topic for its own podcast.
Carroll -
I’m not sure that I agree that an LP would be necessary to complete this study. Especially, when you consider that there is no literature-based guideline of how to “rule-out” a SAH by LP. After all, there is no studies that define what the necessary criteria are. We suspect that if there are no RBCs in tube #4 adn no xanthochromia, then there must be no bleed. Additionally, the CTA-brain does nothing to evaluate for SAH – it only evaluates for the presence of an aneurysm (which we presume is the source of the bleed). I think that lack of the bodies is a reasonable approach for such a disease. The only SAH’s that we care about are the sentinal bleeds. No one cares about the SAH that comes in comatose – there is nothing we can do for them. The only ones that anyone cares about is the ones that are intervenable for short-term outcomes. It’s the same reasoning I use for a lot of things we do in medicine – such as all of this scanning for small PEs… Where’s the bodies?
Schauer
And I forgot to comment on the hyptensive emergency…
I also think that this is a term that needs to be done away with. The hypertension needs to be viewed as a SYMPTOM not a disease. Essential hypertension is a symptom of vascular stiffening that occurs with age and decreased compliance. The elevated blood pressure of a “hypertensive emergency” is therefore not a disease in and of itself that needs to be addressed. It is a symptom of the underlying cause – i.e. SAH, dissection, CHF. Also, there is no literature supporting any type of hypertensive screening – there is no science supporting any claim that a bumped creatnine, RBCs in the urine, elevated cardiac enzymes, or a headache are at all linked to acute blood pressure changes. Things like creatnine changes are long-term prognosticators, not short-term assessments. QED, there is no such thing as a hypertensive emergency – rather there is such a thing as a medical emergency that may have hypertension as a symptom.