Sunday 25 April 2021

Diagnostic Accuracy of the HINTS Exam in an Emergency Department: A Retrospective Chart Review

The HINTS exam for evaluation of patients with acute vestibular syndrome (AVS) has been all the rage over the past decade. Originally described in 2009, it reported phenomenal accuracy at differentiating peripheral from central causes of vertigo. However, the performance of the test has mostly been validated by neurologists and otolaryngologist.

To qualify for a HINTS exam, patients must have AVS characterized by continuous dizziness and ongoing in the ED, nystagmus, and gait unsteadiness.

This retrospective chart review looked at how often a HINTS exam was appropriately performed and the sensitivity & specificity of the test at this single ED in Ontario. The authors point out that doctors did not get any specific training on the HINTS exam prior to this review.

A total of 2,309 patients presented with a primary complaint of vertigo or dizziness. Almost 20% got a HINTS exam! However almost all (96.9%) of them were performed inappropriately as it was not clinically indicated. In addition, many patients got both a Dix-Hallpike test and a HINTS exam… which should never happen as they are “intended in mutually exclusive patient populations.” 

In the end, the sensitivity & specificity of the test as performed was terrible.

What are we to conclude?

At this single ED, the HINTS exam performed poorly in untrained doctors who did the exam on the wrong patients.

They conclude that “additional training of emergency physicians may be required.” This is quite an understatement.

Of course, this study is far from perfect, but it serves as a cautionary tale.

This is not the only study suggesting a problem. A systematic review and metanalysis in 2020 concluded that “the HINTS exam, when used in isolation by emergency physicians has not been shown to be sufficiently accurate to rule out a stroke in those presenting with AVS.”

I do not think the HINTS exam is overly simple. Nor did I believe there are enough cases of vestibular stroke for an average emergency physician to develop and maintain proficiency with the exam. But perhaps I’m wrong… there are some very clever doctors who disagree with me.

Until such time we get further evidence, we should not be giddy over the HINTS exam.

 (** Addendum 28/4/2021: Please see comment section for feedback on my summary from Dr Peter Johns. He is a Canadian Emergency physician and vertigo guru. He wrote the book on vertigo... literally wrote the chapter in Tintinalli)


Dmitriew C, Regis A, Bodunde O, et al. Diagnostic Accuracy of the HINTS Exam in an Emergency Department: A Retrospective Chart Review. Acad Emerg Med 2021;28:387-393. [link to free full text]



1 comment:

  1. I reached out to Dr Peter Johns to provide some feedback on this post on the HINTS exam. Peter is quite the expert on vertigo and the HINTS exam. He has some great instructional YouTube videos and co-wrote the most recent chapter on vertigo in the 9th edition of Tintinalli. He has agreed for me to share his comments below.

    Hi Brian,

    I have a lot of thoughts.

    1. The recent studies by Ohle are to serve as a baseline so we can measure the effect of educational interventions. I am a co-investigator for a study of BPPV which is just starting up.
    2. ED docs are currently terrible at all aspect of vertigo. Try and find a study that says we are good at it. I've attached another study by Ohle (and I am a co-author on this one) that shows we can't even figure out who to do the Dix-Hallpike test on.
    3. The neurologists and ENT at my local hospital, know less about HINTS than I do. (except for two, who have a special interest in vertigo).
    4. I see you're the local ultrasound co-ordinator. I'm old enough to remember when there was zero evidence that ED docs could perform it. Tell me, how did that change? Well, to a certain extent, that's what we have to do with vertigo.
    5. You probably have noticed that not all ED docs of a certain age were able to embrace u/s, but much better uptake in the younger docs. Likely the same with vertigo. Trying to correct all the misinformation we have been taught for decades isn't easy.
    6. The hardest part of HINTS to perform and interpret is the head impulse test (HIT). In order to do it, you have to rapidly turn the head from 20 degrees off center rapidly back to center, and observe the catch up saccade that will be seen in vestibular neuritis. Given what we routinely expect ED docs to learn, I don't think this is a stretch.
    7. In the ED, vestibular neuritis is much more common than a cerebellar stroke masquerading as vestibular neuritis. So we have the opportunity to learn how to perform and interpret the HIT much more frequently than most neurologists and ENT.
    8. An often overlooked fact, HINTS is not the first defense against a dizzy stroke. First line of defense is looking for central features as in the central part of my algorithm in Tintinalli's 9e. Most dizzy strokes will have one of these features, and therefore the presentation will not be consistent with vestibular neuritis and the HINTS exam will not be indicated, as a search for a central cause is now mandated.
    9. Every single test, procedure and decision rule can be improperly performed, or interpreted improperly. That doesn't mean it can't be taught properly. Vanni's study showed that ED docs could be taught how to diagnose both positive and horizontal canal BPPV, and perform the HIT. (study attached). Admittedly, this was a lot of training, but how much training do your trainees get in u/s?
    10. In order to roll out training in vertigo we are going to need to get people who are local vertigo champions to get involved. Currently they are few and far between. But what is the alternative? Continue to not properly assess and treat BPPV, and continue to miss dizzy strokes? MRI everyone? (misses 10-20% of strokes in first 24 hours).

    So Robert Ohle is also the author of the 2020 systemic review.
    Robert was a resident in our program before moving to Sudbury, and I think it's fair to say that I inspired him to study vertigo. And he has been kind enough to include me as a co-investigator in his research.

    His contention (as is mine) is that we are bad at vertigo right now, and need to improve.
    That's why we are conducting educational research about vertigo, starting with the low hanging fruit of BPPV.