NeuroVoices: Anne-Maree Kelly, MD, on Evaluating and Improving Care for Migraine in the ED
The director of the Joseph Epstein Center for Emergency Medicine Research provided context on a study that showed wide variation in EDs diagnosis, treatment, and management of headache.
Anne Kelly, MD
To better describe the epidemiology of nontraumatic headache in emergency departments (ED), an international, multicenter, cross-sectional study was conducted by Anne-Maree Kelly, MD, director, Joseph Epstein Center for Emergency Medicine Research, and colleagues. The analysis, which spanned over 67 hospitals in 10 different countries, included data on the demographics, clinical assessment, investigation, treatment, and outcomes of patients presenting to EDs with this condition.
The findings highlighted the challenges faced when diagnosing and managing headache in the ED setting. Across centers, more than 30 different diagnoses were made, and 27.2% of cases were considered severe. Although most patients were treated with simple analgesics, there was still wide variation observed in the types of treatments used, as well as the neuroimaging tools utilized. CT scan usage ranged from 15.9%-75% across different countries.
According to Kelly, the results serve as a good reminder to institutions to self-evaluate their practices and see how they stack up against others. On a new iteration of NeuroVoices, she provided context on how the study came about, the biggest take-home points, and why progressive change will only come when international and national bodies can agree on consensus guidelines.
NeurologyLive®: Can you discuss the motivation behind your study?
Anne-Maree Kelly, MD: The study was the brainchild of my colleague, professor Kevin Chu, MBBS, MS, from Brisbane. In emergency departments, we see patients who have symptoms and undifferentiated causes for those symptoms, rather than a diagnostic label or even a working diagnostic label. Our role is a little different than that of a neurologist. We try to make a diagnosis where we can, but there’s a lot of risk stratification going on. How high risk is this patient? What do we need to do here and what can be done somewhere else? The group has done some previous research on dyspnea and shortness of breath and had some unusual findings which rocked the cardiology/respiratory world. Those patients showed shortness of breath pages, lungs, and hearts. We found a lot of other causes and other things to think about.
Kevin came to the group and said, “How about we look at headache?” He has had a long interest in headache, particularly thunderclap headache, and so, he brought his experience and awareness of the variation in guidelines about imaging and treatments and said, “Well, why don’t we try to understand more, because if we understand more, maybe it will improve what we do.” And also improve our communication with our colleagues, particularly neurologists, neurosurgeons, etc. Understanding the distribution of diagnosis and how we treat people is the first step in that process. We reached out to our network of people who’ve been involved before and were pleased that so many other health services jumped on board from so many places around the world.
What were some of the topline findings?
We had lots of patients—over 4500 patients. In fact, we just had another cohort of patients provided to us from Columbia, so we’ll have a South American angle that we can include. The top-line findings were that about 1 in about every 100 emergency department patients come in with a headache as a main symptom. It’s not an overwhelmingly common problem, but it’s an everyday problem in an emergency department.
We found wildly varying rates of advanced imaging, particularly CT scanning, ranging from about 1 in every 4 patients, to 3 in every 4 patients, and that variation appears to be geographical. Just as a side note, we’ve done some more analyses and it seems to not correlate well with positive findings. We found very low rates of LP and fundoscopies. The variation in drug use was amazing, and particularly concerning was the high rate of opioids and a big variation in practice about migraine management, which let’s just say, didn’t seem to comply with any guidelines.
What international guidelines are currently in place that address neuroimaging and treatment of nontraumatic headache in the ED setting?
Let’s take the imaging and medication parts as 2 bits. Current guidelines for imaging focus on the acute headache and particularly the thunderclap headache. They’re quite focused on subarachnoid and to a lesser extent, the so-called reflex, which everybody will be aware of. That’s age, pregnancy, postpartum, etc., the snoop 10 criteria. They tend to be quite local, rather than international, but they are broadly similarly.
One of the weaknesses of that approach, however, is that the red flags were derived from people who have been identified as having serious secondary causes and are not necessarily validated on a broad ED population. Because a broad ED population is quite different from any hospital population. The treatment guidelines are diverse and are local or at-best, national, if they exist at all. Many countries and many hospitals have migraine guidelines which are not consistent with each other. But there’s also interdisciplinary differences in guidelines about migraine. For example, the neurologists and the headache society make different recommendations to emergency physicians’ experts particularly with respect to migraine management. That said, everyone accepts that opiates are bad, but unfortunately, they are still used.
What are the major take-home points from your research? Should institutions reflect on how they operate?
One of the things this research highlights and hopefully starts dialogue between groups, is the challenge of the assessment and management of headache in an emergency department. We found more than 30 defined diagnoses for this condition, and only a proportion of them have to do with intracranial problems. There were dental problems, sinusitis, hyponatremia, carbon monoxide toxicity, to name a few. We need to increase the understanding that it’s not quite as simple as perhaps a neurologist or neurosurgeon thinks it is, because they are only thinking in this box, rather than thinking more broadly.
Our research does confirm that benign causes are by far the majority, which was reassuring, and that’s consistent with around the world. That is something that we needed to reassure. It does show that organizations should be doing some auditing about what they’re doing, so that they understand where their CT rate fits among others. They also want to look at misdiagnosis, how much opiates they’re using, and whether those in the migraine subgroups are receiving treatments that are at least within the ballpark of recommended treatments.
I suppose one of the problems is that the recommended treatment for mild migraine, which is nonsteroidals, paracetamol, etc., doesn’t apply to the ED population because they only came because that didn’t work. Recommending that as the first-line treatment doesn’t make sense. We should be jumping that level and going to another level, but the guidelines that are in existence still talk about giving more of that, which the pain continues.
One of the challenges this also brings up is the question of interdisciplinary recommendations about standard of care. Fundoscopy is a good example. We found few people were doing fundoscopy in the ED. That’s partly because Eds tend to have the ancient, handheld ophthalmoscope, which is almost impossible to get a good look at anything, particularly an ED condition. ED physicians know it doesn’t work, but they don’t have access to the technologies that neurologists have, like retinal photography, etc. The pandemic has made that worse, because obviously using a handheld ophthalmoscope in full PPE is damn near impossible. If retinal examination and fundoscopy is important, we need to rethink how we can change our paradigm to make that happen. I’m working with a research group here to start doing some investigation of it. I hope this gives you a flavor of where I hope this might take us.
Transcript edited for clarity.
