Migraine in the Emergency Department Setting: A Chat With Anne-Maree Kelly

By Lincoln Tracy | May 11, 2022 | Posted in

“Why don’t we personalize care a bit? Why don’t we empower patients to self-manage better but also empower them to communicate what things work for them and what things don’t work for them? We could improve their management of migraine, but also decrease ED visits and give people a better quality of life. It can be a win-win-win situation.

Anne-Maree Kelly, MD, is a senior emergency physician and Director of the Joseph Epstein Centre for Emergency Medicine Research at Western Health in Melbourne; Professorial Fellow at the University of Melbourne; and an adjunct professor at the Australian Centre for Health Law Research, Queensland University of Technology.

Her clinical and research interests are wide ranging, with a focus on practical clinical research, improving quality of care in emergency departments, and legal medicine.

Kelly spoke recently with Lincoln Tracy, a research fellow and freelance writer from Melbourne, Australia, about her journey to emergency medicine and the challenges of managing migraine and other headache disorders in an emergency department (ED) setting. Below is a condensed and edited transcript of their conversation. Follow Kelly on Twitter @kellyam_jec  

What was your path to emergency medicine?

Emergency medicine in Australia started in the 1980s, around the time I was starting my career. I had completed my basic medical training and was at the point of choosing a specialty and applying for accredited training positions. I had started an application for surgical training but was asked to fill in at a rural emergency department and I loved it; the variation of emergency medicine really suited me. I ended up changing track and was one of the first doctors to complete the then-new emergency medicine training program.

Anne-Maree Kelly

Anne-Maree Kelly

I was very lucky to be in the right place at the right time, and opportunities came my way. I ended up being appointed to the first academic job in emergency medicine anywhere in Australia or New Zealand. I was a relatively new consultant working at the Queen Elizabeth Hospital in Adelaide and saw an advertisement for an academic position in Christchurch. I thought I’d put my hand up to let people know I was interested but I wasn’t expecting anything; I thought I was too junior to get it. I was lucky enough to get an interview, and then they offered me the job. So I moved countries and spent about three years in New Zealand before moving to Melbourne.

How did your interest in research develop?

As I progressed in emergency medicine, I had questions about my practice – about what we really knew about our patients, and about the evidence for the treatments and investigations we used. When I couldn’t find the answers, I started developing my own research to answer them.

I have a broad range of research interests, from acute cardiology, blood gas analysis, migraine and headache through to shortness of breath – the list goes on. Very early in my career I became aware there were a lot of people who came to the ED with headache. Many of these headaches were labeled as migraine, for one reason or another, but we didn’t do a very good job treating these people. We also often saw the same people coming back to the ED for treatment again and again. At that time, there was a tendency to use opiates, which I didn’t think helped patients, and the debate about meperidine addiction was well and truly happening.

I started to wonder what I could find out about the migraine population who come to EDs – what I could learn about the evidence for what we should do.

I was concerned to find that most of the research was either about patients admitted to the hospital or patients treated in neurologists’ offices. There was not a lot of evidence about the subset of patients who present to the ED. I knew many had taken their usual medications without success and were looking for something else. When I didn’t find the answers to my questions, or only found partial answers, I started asking my own questions and started to answer them.

How significant is migraine in emergency departments?

I must confess that “migraine” as a term is used a bit loosely in EDs, and not always according to the specific diagnostic criteria neurologists use. Consequently, “migraine” in EDs probably overlaps a bit with other forms of recurrent benign headache.

In our recently published study of headache in the ED, which involved 67 hospitals in 10 countries, we found that headache accounted for about 1% of ED presentations. About a third of those were coded as migraine. In my current practice in Melbourne, we would see somebody with a migraine-type headache at least once a day.

Some patients are straightforward to manage. They have their typical headache, but their medication just hasn’t worked, and they want something a bit stronger that they don’t have access to. Others are less straightforward. They’ve got some atypical features, or they’ve got the worst migraine they’ve ever had in their life. This poses diagnostic challenges. Do these patients need imaging or another test to ensure we don’t miss an important alternative diagnosis?

There are therapeutic challenges as well. Patients have drugs that they like to be treated with. Migraine is interesting to consider with respect to individual responses to medication. There are people who respond best to treatments that the evidence says are not the most effective – and I’m not just talking about opiates. In my experience, dopamine antagonists are underrated. Choosing the right medication, particularly in busy EDs that are overcrowded, can be challenging.

You have published several papers from the Headache in Emergency Departments, or HEAD study, which is an international collaboration. How did this collaboration start?

I have always been a collaborative researcher. I’ve always felt we are more powerful, both in terms of thinking and data, when we work together. But it’s not always easy to get people to collaborate in the competitive academic world, and emergency medicine research is terribly underfunded.

About 10 years ago I was introduced to Professor Said Laribi from France at an emergency medicine conference, and he was also a fan of collaborative research. We got to chatting and started talking about whether we could collect data together to have more powerful data. We approached our networks in EDs across Europe, Australasia, and Southeast Asia, which led to a project on shortness of breath; this project collected data on 3,500 patients. The insights from that study, and its 12 papers, have started influencing guidelines and translating into practice change.

After the success of that work, we looked at each other and asked, “Shall we have a go at seeing what we can do with headache?” Because there was so much enthusiasm, we got more sites the second time around, and more than 4,500 patients in the headache study. Then, about a year ago, I was approached by a Colombian hospital that has now contributed 750 patients from South America. Now we have over 5,000 patients and four continents represented.

The South American data have helped us get enough data to, for example, analyze whether pregnancy is a true red flag in headache. Our data suggest it probably isn’t. We’re also planning on challenging some of the SNNOOP10 criteria [criteria for secondary headache disorders] – whether those criteria are valid in the ED headache population. So far, this study has four publications, with more on the way. It’s very exciting.

Were there any surprising findings from the migraine-specific component of the HEAD study?

One thing that surprised us was that 25% of patients presenting to an ED with migraine had a CT scan. This is an important finding, because if we expose migraine sufferers to radiation every time they come to the ED with a migraine, that’s probably going to have some downstream negative effects.

Finding the best balance of risk and benefit can be challenging. It needs to consider a person’s normal symptomatology, and how that fits with the overall clinical features this time around. In my opinion, it should also involve some shared decision making with the patient about the risks and benefits of treatment, observation, and reassessment. Also, it’s important to ask questions like, “How many CTs have you had overall?” and take that into account before you order the next one.

Your recent paper also expressed concern about the use of opiates to treat migraines. Can you elaborate on those concerns?

I think that the use of opioids for migraine came about, in part, through a lack of evidence about what really worked in the migraine population and a rising awareness of the undertreatment of pain in EDs. That led to an uptick in the use of opiates, which can be appropriate for things like severe abdominal pain and fractures but are less appropriate, and best avoided, in recurrent painful conditions such as migraine.

At the time, there was little data about what worked to treat headache in the ED population, especially migraine. There was very little published, so opioids filled the void. Now we’re having the trouble of undoing that.

Part of dealing with the challenge is managing patient expectations, but part of it is about changing ED practices and saying that it’s not okay to give an opiate for headache as a first choice. There may be situations where a small dose of an opiate may be necessary to get pain relief – if there is an intracranial bleed, for example – but an opiate should never be a first-choice agent when the likely diagnosis is a benign headache.

Within the field of migraine research or clinical practice, is there something we don’t currently know that you hope we will have an answer for in five to 10 years?

Obviously, there’s the hope that therapeutics will advance. We also desperately need good data about practices that raise concerns among emergency physicians, such as the use of ketamine and propofol in EDs for treatment of migraine, which have significant potential for dependence.

I’d like to see us do something a little different in clinical practice, taking the lead from the asthma action plan research. For people for whom migraine is a recurrent problem, could they have an action plan with their family doctor or neurologist that addresses questions like, “What should I do when I first get my headache? What medication should I take? What activities should I do or not do? When should I go to an ED?” Then, when they come to the ED, they could bring the information about what they’ve had in the past and what works for them, so we can have a personalized plan and give them something that works the first time they visit the ED.

One of the findings of our big study was that about 40% of people ended up getting rescue medication because they didn’t receive the optimal abortive therapy the first time they sought medical attention. Having an action plan, which would also include follow-ups and a consideration of corticosteroids to prevent recurrent headache, would be interesting.

The evidence from the asthma action plan approach is that it works. So, I’m thinking, Why don’t we personalize care a bit? Why don’t we empower patients to self-manage better but also empower them to communicate what things work for them and what things don’t work for them? We could improve their management of migraine, but also decrease ED visits and give people a better quality of life. It can be a win-win-win situation.

Lincoln Tracy, PhD, is a researcher and freelance writer based in Melbourne, Australia. Follow him on Twitter @lincolntracy

Related reading:

Migraine in the emergency department: A prospective multinational study of patient characteristics, management, and outcomes. Wijeratne et al. Neuroepidemiology. 2022;56(1):32-40

Sign Up For An MSC Newsletter

Dr. Lincoln Tracy is a researcher and freelance writer from Melbourne, Australia. As a researcher, he uses data from an international clinical quality registry to explore burn injuries in Australia and New Zealand. As a freelance writer, he turns basic, translational, and clinical research into high-quality news, features, interviews, meeting reports, and podcasts. As a person, he is one half of one of two sets of twins in his family. Follow him on Twitter @lincolntracy.



  • Content Types

  • Paper Categories

  • Paper Dates

Recent News

Recent Papers


No event found!