Date: April 12, 2023
Reference: Edlow et al.Guidelines for Reasonable and appropriate care in the emergency department 3 (GRACE-3): Acute dizziness and vertigo in the emergency department. AEM May 2023
Guest Skeptic: Dr. Jonathon Edlow has practiced emergency medicine for nearly 40 years and is a Professor of Emergency Medicine at Harvard Medical School. His academic interest is avoiding the misdiagnosis of patients with neurological emergencies.
This is an SGEM Xtra and #SGEMHOP combined. We have reviewed the previous two GRACE guidelines published by the Society for Academic Emergency Medicine (SAEM). GRACE stands for Guidelines for Reasonable and Appropriate Care in the Emergency Department.
This is the third GRACE project. GRACE-1 tackled the common issue of recurrent low risk chest pain (SGEM#337). This contrasts with other guidelines that only looked at a single emergency department presentation for chest pain. And GRACE-2 was about low-risk recurrent abdominal pain (SGEM#367).
Now we come to GRACE-3. This time it was not a recurrent condition like chest pain or abdominal pain but rather acute vertigo or dizziness. The objective of the GRACE-3 guideline is to provide an evidence-based framework intended to support patients, clinicians, and other health-care professionals in their decisions about the evaluation and management of adult ED patients with acute dizziness who do not have an obvious central cause with frank neurological findings or an obvious general medical one.
The population covered by these guidelines are adult patients presenting to the ED with acute dizziness or vertigo of less than two weeks. Let us clarify some terms because it is often not clear when what people mean by dizziness and vertigo. I tend to describe dizziness as light-headedness, unsteadiness, motion intolerance, imbalance, floating, or a tilting sensation.
Dizziness in GRACE-3 was defined as the sensation of disturbed or impaired spatial orientation without a false or distorted sense of motion (Barany Society). Vertigo in GRACE-3 was defined as the sensation of self-motion (of head or body) when no self-motion is occurring, or the sensation of distorted self-motion during an otherwise normal head movement. The problem is that research shows that patients often use multiple descriptors simultaneously or change their main descriptor if asked again less than 10 minutes later. So, although your concept is exactly what has been taught for decades, data from the last decade and a half show that it’s simply not true.
The author group came up with 15 evidence-based recommendations based on the timing and triggers of the dizziness. Let’s go through those recommendations with the first one being an overarching one.
15 Recommendations from GRACE-3
Recommendation 1: Emergency clinicians should receive training in bedside physical examination techniques for patients with the AVS (HINTS) and diagnostic and therapeutic maneuvers for BPPV (Dix-Hallpike test and Epley maneuver), since untrained ED physicians do not reliably apply or accurately interpret results of this bedside eye movement examination. [Ungraded Good Practice Statement]
- HINTS stands for Head Impulse, Nystagmus, and Test of Skew. It was initially touted as a highly sensitive, specific marker for cerebellar stroke in the ED. However, study results have been mixed and there is ongoing debate about how to train for and utilize this examination in the acute care setting.
A 2021 study  reported that EM physicians could be trained on the HINTS and it gave a sensitivity of 97% for central vertigo (SGEM#376). However, a SRMA  that included EM physicians showed less impressive results with a sensitivity of 83. The authors of the review felt the use of HINTS by EM physicians had “not been shown to be sufficiently accurate to rule out a stroke.” (SGEM#310).
There is a clear disconnect between what is possible for emergency physicians to do and what we currently do in real life. It is possible that emergency physicians can learn to use these techniques (not only the HINTS exam, but also bedside maneuvers to diagnose and treat BPPV). This is consistent with my own anecdotal experience. What is equally clear is that without some sort of training or intentional activity to learn it, we do a pretty bad job in routine practice. Dr. Peter Johns has some great videos on YouTube to show clinicians how to do the HINTS exam. I don’t know this for sure, but I suspect that Peter, like me, picked this up by active learning, not by attending a training session.
GRACE-3 goes on to help distinguish central from peripheral causes in patients with the acute vestibular syndrome (AVS). AVS is a clinical syndrome of acute-onset continuous dizziness lasting days to weeks and generally includes features suggestive of new, ongoing vestibular system dysfunction (e.g., nausea and vomiting, nystagmus, and postural instability). In the ED, patients are symptomatic even at rest, and exacerbation from head movement or position change is typical. This is different from episodic syndromes that can be spontaneous or triggered.
Recommendation 2: In adult ED patients with AVS with nystagmus, we recommend routine use of the 3-component head impulse, nystagmus, test of skew (HINTS) exam for clinicians trained in its use* to distinguish between central (stroke) and peripheral (inner ear, usually vestibular neuritis) diagnoses. (Strong recommendation, FOR) [High certainty of evidence]
This is very important. There was a large chart review of EM physicians that reported EM physicians misapplied the HINTS exam 97% of the time .
Recommendation 3: In adult ED patients with AVS with nystagmus, we suggest assessing hearing at the bedside by finger rub to identify new unilateral hearing loss as an additional criterion to aid in the identification of stroke, even if the 3-component HINTS exam result suggests a peripheral vestibular diagnosis. (Conditional recommendation, FOR) [Moderate certainty of evidence]
Recommendation 4: In adult ED patients with AVS without nystagmus, we suggest assessing severity of gait unsteadiness to help distinguish between central (stroke) and peripheral (inner ear, usually vestibular neuritis) diagnoses. (Conditional recommendation, FOR) [Moderate certainty of evidence]
This is a hugely important point. One should be very hesitant to make a diagnosis of a benign, peripheral cause of dizziness in a patient who cannot walk independently. Plus, in terms of disposition in general, it’s not a great idea to discharge someone who cannot walk (assuming they could walk before). This simple test is often admitted.
Recommendation 5: In adult ED patients with AVS with or without nystagmus, we recommend against routine use of non-contrast computed tomography of the brain (CT) or (CT angiography [CTA]) to help distinguish between central (stroke) and peripheral (inner ear, usually vestibular neuritis) diagnoses. (Strong recommendation, AGAINST, see Implementation Considerations) [High certainty of evidence]
That goes against the EM physician alphabet. A-B-C-T. We like to send these patients to the “donut of truth”. What this recommendation says strongly is don’t use these imaging modalities (CT or CTA) to distinguish between central and peripheral vertigo.
CT is a horrible test in this setting. Of course, non-contrast CT is a bad test for any hyperacute stroke, but it’s even worse for posterior circulation strokes and even worse still for those posterior circulation strokes presenting as isolated dizziness. This is why we made a strong recommendation against the routine use of CT or CTA based upon high certainty of evidence.
Recommendation 6: In adult ED patients with AVS with or without nystagmus, if a clinician trained in use of HINTS is available, we recommend against routine use of magnetic resonance imaging of the brain (MRI) or cerebral vasculature (MRI angiography [MRA]) as the first-line diagnostic test (prior to physical examination) to help distinguish between central (stroke) and peripheral (inner ear, usually vestibular neuritis) diagnoses. (Strong recommendation, AGAINST, see Implementation Considerations) [High certainty of evidence]
The part about not getting an MRI is a little easier because in many places it can be difficult to get these on an urgent basis. The important part is to get clinicians adequately trained in the use of HINTS.
Agree that it’s easier when it’s hard to get but it can be misleading if you can get it, do get an MRI and it’s normal. In the first 48 hours of an acute stroke that presents as isolated dizziness, MRI only has approximately 80% sensitivity.
Recommendation 7: In adult ED patients with AVS and central or equivocal HINTS results, we recommend use of stroke protocol MRI (with diffusion-weighted images [DWI] and MRA) to further help distinguish between central (stroke) and peripheral (inner ear, usually vestibular neuritis) diagnoses. (Strong recommendation FOR, see Implementation Considerations regarding timing of MRI) [High certainty of evidence]
Recommendation 8: In adult ED patients with spontaneous episodic vestibular syndrome, the writing committee believes that routine use of a detailed history and physical exam with emphasis on cranial nerves including visual fields, eye movements, limb coordination, and gait assessment helps to distinguish between central (TIA) and peripheral (vestibular migraine, Menière disease) diagnoses. [Ungraded good practice statement]
This seems like a motherhood and apple pie statement. Take a good history followed by a directed physical exam. As a reminder, Spontaneous episodic vestibular syndrome (s-EVS) is a clinical syndrome of transient dizziness usually lasting minutes to hours and generally includes features suggestive of temporary, short-lived vestibular system dysfunction during attacks. There may be a history of recurrent attacks, but patients may initially present after or during a first attack. There are no clear triggers for these attacks, although symptoms may be exacerbated by head movement or position change during an attack. These patients are generally asymptomatic at rest.
Some patients with longer duration episodes may have symptoms on presentation to the ED; in this situation, one would approach as an AVS and the true episodic nature may only be apparent in retrospect. Conceptually, this is no different from managing a patient presenting with focal neurological symptoms as stroke even though if the symptoms later spontaneously resolve and imaging is negative, in retrospect, one might diagnose a transient ischemic attack (TIA). The major differential diagnosis here is posterior circulation TIA vs vestibular migraine. The former is the dangerous diagnosis but the latter is orders of magnitude more common and not something that we tend to think about or diagnose in the ED.
Recommendation 9: In adult ED patients with spontaneous episodic vestibular syndrome, we recommend against routine use of CT to help distinguish between central (TIA) and peripheral (vestibular migraine, Menière disease) diagnoses. (Strong recommendation, AGAINST) [Moderate certainty of evidence]
Another strong recommendation against getting a CT scan! If CT is bad for stroke, it’s not surprising that it’s bad for TIA.
Recommendation 10: In adult ED patients with spontaneous episodic vestibular syndrome and concern for TIA, we suggest use of CTA or MRA of the head and neck to rule out posterior circulation vascular pathology (Conditional recommendation, FOR) [Moderate certainty of evidence]
So, we can get CTAs or MRAs for some patients based upon moderate evidence. If you are thinking TIA, then vascular imaging is important, same as with an anterior circulation TIA. Patients with TIA who decompensate early (ie, have a stroke), tend to be those with large vessel disease.
Recommendation 11: In adult ED patients with triggered episodic vestibular syndrome, we recommend routine use of the Dix-Hallpike test to diagnose posterior canal BPPV (Strong recommendation, FOR) [Moderate certainty of evidence]
A Triggered episodic vestibular syndrome (t-EVS) is a clinical syndrome of transient dizziness lasting seconds to minutes and generally including features suggestive of temporary, short-lived vestibular system dysfunction (e.g., nausea, nystagmus, and postural instability). There is usually a history of recurrent attacks, but patients may initially present after a first attack.
In cases of t-EVS, there are clear triggers for these attacks, most often movement of the head. These include postural shifts, as when standing up or getting into bed or head motion related to turning over in bed or looking up towards a high shelf. In the ED, these patients are generally asymptomatic at rest, but symptoms can be readily provoked at the bedside by reproducing the patient’s trigger. It’s important to realize that although there are clear positional triggers in patients with BPPV, patients with any cause of dizziness are often worse with changes in position. Take a patient with a cerebellar tumor, put them on a merry-go-round and their dizziness will ge worse. So worsened dizziness with movement is NOT diagnostic of a peripheral cause. This is a very common (and dangerous) misconception.
Dr. Peter Johns has another great YouTube video to show the Dix-Hallpike test.
Recommendation 12: In adult ED patients with triggered episodic vestibular syndrome, we recommend against routine use of CT or CTA (Strong recommendation, AGAINST) [Moderate certainty of evidence]
Another prohibition on using the CT scanner. “Let me count the ways” . . . that CT results in useless and potentially dangerous (because people conclude that there is no central cause).
Recommendation 13: In adult ED patients with triggered episodic vestibular syndrome diagnosed with typical posterior canal BPPV by a positive Dix-Hallpike test with the characteristic nystagmus, we suggest against routine use of MRI or MRA (Conditional recommendation, AGAINST) [Moderate certainty of evidence]
A couple points of clarification: First, learn when to do the Dix-Hallpike. Second, learn HOW to do it and HOW to interpret it. Third, it only takes about 30 seconds to do and can make a diagnosis and disposition really quickly. Finally confirming a benign diagnosis is a great way of ruling out a worrisome diagnosis.
Recommendation 14: In adult ED patients with a clinical diagnosis of vestibular neuritis, we suggest shared decision-making with patients to weigh risks and benefits of short-term steroid treatment for those presenting within three days of symptom onset. (Conditional recommendation, FOR) [Very low certainty of evidence]
We like doing shared decision making in the ED. It gives patients autonomy and agency over their health care. This is even more important when there is little evidence to support one action over another. In this case, the use of steroids for vestibular neuritis.
Recommendation 15: In adult ED patients with posterior canal BPPV diagnosed by a positive Dix-Hallpike test, we recommend the Epley† canalith repositioning maneuver be performed at the time of diagnosis. (Strong recommendation, FOR) [Moderate certainty of evidence]
See Dr. Peter Johns YouTube video doing the Dix-Hallpike test followed with the Epley maneuver.
Those are the 15 recommendations. Dr. Eldow had some final thoughts:
“I think that trying to place patients into a timing and triggers category, rather than “what do you mean dizzy” category is worth trying to do. The word a patient uses is not diagnostically helpful. But if it’s difficult to get the patient into a timing and triggers category, there is the STANDING algorithm, devised by an Italian EP, Simone Vanni, that has an algorithm that is agnostic to timing and triggers category (Vanni; 2017; Frontiers of Neurology). His work is also important because, like the Gerlier study, it shows that EPs can be trained to perform all of these bedside maneuvers accurately including those for BPPV. He has also published data showing that it takes EPs less than 3 minutes to do these bedside tests. Finally, I would stress that unlike many guidelines, this one is to an extent aspirational. We recognize that most EPs do not feel comfortable with some of these tests, especially HINTS (although I’d point out that the BPPV maneuvers have existed for many decades) and that’s why our first recommendation advocates for training.”
The guideline has a figure to help clinicians recognized some of the common errors in the diagnosis of adult ED patients with acute dizziness.
There is also a nice diagnostic algorithm for approaching adult ED patients with acute dizziness.
Dr. Peter Johns
Dr. Peter Johns is a vertigo expert. He has been practicing emergency medicine since 1985 and has been passionate about vertigo education for the last two decades. He co-authored the Vertigo chapter in the current edition of Tintinalli’s emergency medicine textbook and as we have mentioned many times Peter has a YouTube channel about vertigo that has >16,000 subscribers and five million view. He was on SGEM#310 talking about the HINTS exam.
Listen to what Peter thinks about the GRACE-3 guidelines using this LINK.
Keener Contest: Last weeks’ winner was Dr. Dennis Ren. He knew the “Mutant Leader” was the name of the villain who was faster, stronger and seemingly impervious to pain in the classic Batman graphic novel from 1986.
Listen to the SGEM podcast to hear this weeks’ question. If you know the answer, send an email to TheSGEM@gmail.com. The first correct answer will receive a cool skeptical prize.
REMEMBER TO BE SKEPTICAL OF ANYTHING YOU LEARN, EVEN IF YOU HEARD IT ON THE SKEPTICS’ GUIDE TO EMERGENCY MEDICINE.
- DIFFERENTIATING CENTRAL FROM PERIPHERAL CAUSES OF ACUTE VERTIGO IN AN EMERGENCY SETTING WITH THE HINTS, STANDING, AND ABCD2 TESTS: A DIAGNOSTIC COHORT STUDY. Gerlier C, et al. Acad Emerg Med. doi: 10.1111/acem.14337. July 2021. Online ahead of print
- CAN EMERGENCY PHYSICIANS ACCURATELY RULE OUT A CENTRAL CAUSE OF VERTIGO USING THE HINTS EXAMINATION? A SYSTEMATIC REVIEW AND META-ANALYSIS. Ohle R, et al. Acad Emerg Med. 27(9):887-896, September 2021
- DIAGNOSTIC ACCURACY OF THE HINTS EXAM IN AN EMERGENCY DEPARTMENT: A RETROSPECTIVE CHART REVIEW Dmitriew C, et al. Acad Emerg Med. 28(4):387-393, April 2021