Podcast – Headache – Intro to EM

Welcome back to the St. Emlyn’s induction podcast. I’m Iain Beardsell and I’m Simon Carley. Today we’re going to tackle a rather tricky presenting complaint that I think probably worries every emergency physician at some point in their career: the patient who presents with a headache. These are the patients who come in with a headache as their principal symptom. Many of our patients will often get headaches as part of their symptomatology, but this is where the headache is the main feature. So, as ever, we’re going to think about the things that we need to rule out in the ED and a strategy about how we might be able to differentiate between those worrying diagnoses and get to a point where we’re either treating a patient, admitting them to the hospital, or safely discharging them in the knowledge that in all likelihood there isn’t anything seriously wrong with them.

So Simon, what are the major diagnoses you always consider when a patient presents to you with a headache? Thinking like an emergency physician, we’re always going to go for the things that are most serious and are most likely to kill your patient or kill them quickly. The big things that I’m going to be looking for are: does this person have a subarachnoid hemorrhage? Does this person have meningitis? Do they have a tumor or some other weird stuff going on in the brain? Or do they have something called temporal arthritis? Those are the ones which I really don’t want to miss. I don’t want them to come back in a week’s time either very, very unwell or with somebody else to make diagnoses. I want to make those diagnoses on day one.

So what are the characteristics of a patient with a subarachnoid hemorrhage? How can we make that diagnosis or think to ourselves whether a patient needs further investigation? Really severe headaches and the classic is a sudden onset headache. It’s the person who’s walking down the road, minding their own business, when they felt like a man came up behind them and hit them in the back of the head with a baseball bat. Sudden onset, incredibly severe. In reality, not all of them are like that. Many patients just present with very severe headaches, ones that come on over a period of minutes or even a little bit longer. So just patients in whom you’re worried have got an incredibly severe headache, the lone headache, the worst headache of their life. Those are the ones I start to worry about.

Unfortunately, our patients don’t always read the medical textbooks, do they? It is a long time since I’ve seen a patient who described the classic subarachnoid hemorrhage. Over my career, my threshold for investigating these patients has definitely gone down because we do know that a significant number of patients presenting to the ED with a headache will have pathology. There’s quite a few studies out there that show that about 10% of patients who come through the doors of your emergency department saying, “I have a headache and I want help,” will have something which is potentially life-threatening: subarachnoid hemorrhage, tumor, meningitis. I think that’s quite a high-hit rate. Some people will say, “Oh, 90% of them don’t,” but 10% of them do. One in ten. I’m quite worried about that. That’s very different, I think, from the population who present at general practice.

If you’ve been working in GP land before or that’s really the background from which you’re coming and you’re starting work in the ED, remember different patients come to see us in the ED. You could hypothesize that the way access to out-of-hours care has changed over the years would start to even out, but I don’t think that’s true yet. So remember that the patients coming to the ED we need to take seriously. Subarachnoid is up there as one of those worrying diagnoses. They don’t always present classically, we’ve said that. They can present with a Thunderclap headache. For me, “It’s the worst headache of my life” is the sentence that I’m fearful of. As I say, my threshold for investigating, with the first investigation being a CT scan, has very much lowered now. I don’t mind too much if I get a negative CT on a patient who has a severe headache. I think that’s a worthwhile investigation to do. Would you agree?

I think so, yeah. I think if that patient says those magic words, “This is the worst headache of my life,” they almost bought themselves a CT at that point. An early CT scan in these patients is both diagnostic. The earlier you do it, the more diagnostic it is likely to be for the nature of the way that the CT test works. So early CT scans in these patients are incredibly helpful. If you’ve seen a patient who describes the worst headache of their life in your ED and you want to organize them for further investigation, I think we are both saying that CT would be the way to go. In your hospital, that may involve asking a registrar in the department to discuss that with a radiologist or asking a consultant, but make sure that that CT scan happens. I think we both agree that that’s a worthwhile investigation to do first.

Those patients are relatively straightforward with the worst headache of my life. What do you do about those sort of borderline? I just have a bit of a niggle type thing. I think you’re back to taking a really good history and doing a good examination and exploring what’s going on. You’re looking for associated features which might give you a clue for another pathology. Do they have a respiratory tract infection? Have they got a history of migraines? Is this the 17th time that they’ve had this episode? Those are going to kind of reassure you. But for those patients who got a headache which is of a new characteristic, it’s more severe than before, it’s got other potentially associated symptoms or signs, I’m also going to be quite worried about them. Some of those signs are often quite soft, diplopia, but you can’t demonstrate it on examination. A small period of confusion, headaches worse at different times of the day, although not often the classic diurnal variation that you read in the textbooks, something else which makes you feel uncomfortable about the nature of the headache. In the case of meningitis, often accompanied by fever, a period of being generally unwell, those are the other things that point you in the direction of infection.

So, a patient presented with a headache that described their headache characteristics. There’s some on-set severe ones, that’s okay, we’re going to get straight to CT. Just while we’re on that, do all of these patients need a lumbar puncture after their CT scan? That’s an incredibly controversial question at this stage. The practice in the UK at the moment in most centers is that they would go on and have a lumbar puncture. The new evidence which is coming out now would suggest that if you CT these patients early, a lumbar puncture may be unnecessary. It’s a question that, as we stand at this moment in time, on today, I think needs a consultant level conversation between the patient and the admitting teams to decide what to do with that patient. The easiest thing to think of is that at the moment, a lumbar puncture is going to be the way forward, be that in the ED or with your acute meds and colleagues.

Subarachnoid hemorrhage, that’s the first one we’re going to consider life-threatening, serious, we’re going to do our investigations. Meningitis, do you think that’s an obvious diagnosis to make? It can be. You can see the classical picture of somebody coming in with a recent infection, a very high temperature, obvious meningism, a depressed level of consciousness. But again, it’s a bit like the subarachnoid. Most of these patients have not read the books. Many patients will come in with an insidious onset. They will have neck pain, but not necessarily the classic signs of severe rigidity in the neck. They will be uncomfortable with the light, but not without really harsh photophobia. Again, you need to build a picture with these patients. For me, somebody who’s complaining of a headache as their principal symptom, and who’s got a high temperature and any other associated features associated with meningism, I’d be very concerned.

What would you do to try and delay those concerns? Are blood tests enough? A normal white cell count, negative CRP? Was that helpful? There’s plenty of evidence around that those tests are not helpful. They really aren’t, and they can confuse you and make you do the wrong thing. So reaching for the phlebotomy set in the blood test may not be easiest for differentiating whether this is meningitis or something else. The greatest advantage there is you putting the line in for giving the antibiotics for the suspected meningitis in front of you because time is important. But we’re reiterating again not to rely on blood tests, especially when it comes to ruling out these life-threatening diseases, because negative tests don’t necessarily mean the patient hasn’t got the disease.

I think it’s a really good point. I think when people go and see these patients and they’re worried that it might be meningitis, there’s sometimes a caution in, “If I make this diagnosis and I give antibiotics, am I going to look a bit silly later on?” That’s a possibility. You might potentially go and see a patient, and maybe they didn’t have meningitis, maybe they just had a bit of flu and they had a headache and a bit of a temperature, and you gave them some antibiotics, and in a few hours time somebody will go, “Yeah, maybe they didn’t need it.” I’ve got to say, I’m quite happy to have many more conversations like that than the one conversation when somebody comes along and says, “You didn’t give antibiotics to this person who subsequently had meningitis, and they’re really not very well at all now.” Having had one dose of antibiotics or two doses and then being proved it’s not meningitis isn’t a big deal. Obviously, allergies, we check about and all those other things. To miss a patient with meningitis is career-changing and life-changing for both the doctor and the patient.

We’ve got an idea on meningitis, and all the time we’re pointing out here that CT scanning is the key investigation for many of these patients with a headache. Blood tests may not be that useful. Obviously, we mentioned tumors. They’re going to be pretty well spotted on CT scan most of the time, but not absolutely always, and sometimes you may need to do a CT and angiogram or even MR to identify some tumors. They can be quite difficult to pick up, and they often present to the emergency department with fairly non-specific signs. Probably one of the most common reasons we diagnose them is post-fit in a young person. But I reckon we pick up several, well I know we pick up several intracranial tumors every year through the emergency department, so it’s not actually that rare of a diagnosis for us to make.

I agree, I don’t think it is at all. Some of my most memorable and most upsetting patients have been those who’ve, as you say, come in with a first fit and they’ve ended up having a scan and you found something pretty devastating. Throughout all of this, we’re just reiterating that CT scanning is an important investigation. Had we been recording this podcast 15 years ago, I don’t think that’s what we would have been talking about, but the threshold for scanning has undoubtedly gone down. I realize that we’re worried about radiation risk, especially in young people, but in that harm-benefit analysis, with these patients presenting to the emergency department where we’ve said 10% of them can have serious pathology, doing a CT scan to be sure that they haven’t got some of the things we’re talking about, I think is important. The decision to not do a CT scan is a senior decision, I think, so perhaps we’re suggesting your default position should be to do a scan only to be asked not to by a senior clinician who’s reviewed the patient with you.

Subarachnoid hemorrhage, meningitis, tumors, we also mentioned temporal arthritis. I’m quite interested in temporal arthritis for various different reasons, but I think it is one of the diagnoses that you can make in the emergency department where blood tests do help. In all patients I see over the age of 50 or 55, I think 55 in the literature, we sometimes drop it down to 50, who’ve got a headache as a presenting feature. We don’t think it’s one of the other biggies, always taking the ESR and CRP, and if that’s significantly raised, then question whether or not this patient could have temporal arthritis.

Is this the one time where both you and I will suggest that the so-called inflammatory markers are useful? Indeed, although it can get really complicated and a patient who genuinely believes has got temporal arthritis, the inflammatory markers can be normal, but I think it’s a good diagnosis to look for and have in the back of your head. Definitely worth doing. Certainly in patients aged over 55, do an ESR, do a CRP. Again, we know that early treatment can be sight-saving, very important for us to pick up in the motor run. This isn’t the one you want to come back two days later, presenting with visual disturbance when you could have done something about it on the day you saw them.

Interestingly, the last one you talked about isn’t necessarily a life-threatening diagnosis. Actually, on the first day I did emergency medicine, I was taught never to diagnose migraine in the ED. Yet, we were going to spend a few minutes talking about it. I think so, because maybe it’s a personal thing, but I get migraines and they’re fairly unpleasant actually. You and I both agree that one of the best things that we can do in emergency medicine is relieve pain and suffering. If somebody’s got a migraine, which is so severe that it’s brought into the ED, if they have not managed it well at home perhaps, but if they’ve got a very severe headache, maybe there is something we can do for them, which could be helpful.

There are a number of strategies around, and I don’t think it’s completely clear what the best way of treating migraine is yet, but a number of strategies involving rehydration, involving 5HT3 drugs, involving antiemetics, and simple analgesics, which we can do to assist these patients. But more than that, I think it still goes back to those original diagnoses. If somebody comes into the department and they say, “I think I’m having my first migraine,” I just want someone to take a really good look at them. If it’s their 15th attack of migraine fine, we’re just doing analgesics and symptom control, but if somebody else has made a diagnosis of a severe migraine, it’s the first one they’ve ever had, and they’re 65 years old, hang on a minute, that’s not right. I want someone senior to go and see them.

So there’s two sides to that aren’t there. There is the patient like yourself, and I promise if you ever have a migraine in my company, I will look after you, that we need to just make sure we relieve their symptoms, and we will always come back to people who come to us asking for help. More often than not, they do want to know what’s wrong with them, but first of all, they want somebody to take the pain away. To come to the emergency department, which we’ve always said is a busy place, no one wants to be there, with your headache, it must be pretty bad, so never forget the analgesics here. If you get through this diagnostic pathway, and it is a migraine, then explore something that you can give them to help. You mustn’t ever discount the idea that the person who says, “I think I’m having my first migraine,” or you take the handover from an ambulance member who says, “Oh, this is the first migraine, it sounds just like a migraine.” Always go back and consider the life-threatening or life-changing diagnoses, so those of subarachnoid hemorrhage, meningitis, tumors, and other funny stuff going on intracranially, and then temporal arthritis. We have to actively rule those out before we can make the decision to treat them for something that isn’t as serious.

I agree with you that, and it’s further evidence that these patients are actually potentially quite complex, both in terms of how we investigate them and whether we make a reasonable discharge decision. I think it is one of those diagnoses where, as a junior doctor in the emergency department, I would certainly get one of my senior colleagues to see the patient, discuss the patient with them, and to explore how they would rationally investigate them. Only by talking through that process do you get inside the mind of an emergency physician about how they think, and that’s all part of the learning process.

We’ve just covered very briefly some of the things that we want you to think about when you see the patient with a headache as their presenting symptom in the emergency department. Of course, there’s an associated blog post on the website with some other resources that you can use, including an excellent presentation by Sarah Robinson, one of my colleagues from Southampton. We hope you’re getting something from all of these educational sessions. We’d love to hear from you, please do get in touch. But for now, keep enjoying your emergency medicine, and we’ll talk to you again soon. Take care. Have fun!