The Journal of Emergency Medicine, Vol. 50, No. 4, pp.

617–628, 2016
Copyright Ó 2016 Elsevier Inc.
Printed in the USA. All rights reserved
0736-4679/$ - see front matter

http://dx.doi.org/10.1016/j.jemermed.2015.10.040

Best Clinical
Practice

USING THE PHYSICAL EXAMINATION TO DIAGNOSE PATIENTS WITH ACUTE

DIZZINESS AND VERTIGO

Jonathan A. Edlow, MD* and David Newman-Toker, MD, PHD†

*Department of Emergency Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts and †The
Johns Hopkins Hospital, Department of Neurology, The Johns Hopkins University School of Medicine, Bloomberg School of Public Health,
The Johns Hopkins University, Baltimore, Maryland
Reprint Address: Jonathan A. Edlow, MD, Department of Emergency Medicine, Beth Israel Deaconess Medical Center, 1 Deaconess Place,
CC-2, Boston, MA 02215

, Abstract—Background: Emergency department (ED) (BPPV), but some have posterior fossa mass lesions. Specific
patients who present with acute dizziness or vertigo can be positional tests to provoke nystagmus can confirm a BPPV
challenging to diagnose. Roughly half have general medical diagnosis at the bedside, enabling immediate curative
disorders that are usually apparent from the context, associ- therapy, or indicate the need for imaging. Conclusions:
ated symptoms, or initial laboratory tests. The rest include a Emergency physicians can effectively use the physical exam-
mix of common inner ear disorders and uncommon neuro- ination to make a specific diagnosis in patients with acute
logic ones, particularly vertebrobasilar strokes or posterior dizziness or vertigo. They must understand the limitations
fossa mass lesions. In these latter cases, misdiagnosis can of brain imaging. This may reduce misdiagnosis of serious
lead to serious adverse consequences for patients. Objective: central causes of dizziness, including posterior circulation
Our aim was to assist emergency physicians to use the phys- stroke and posterior fossa mass lesions, and improve
ical examination effectively to make a specific diagnosis in resource utilization. Ó 2016 Elsevier Inc.
patients with acute dizziness or vertigo. Discussion: Recent
evidence indicates that the physical examination can help , Keywords—dizziness; vertigo; physical examination;
physicians accurately discriminate between benign inner posterior circulation stroke; misdiagnosis; BPPV
ear conditions and dangerous central ones, enabling correct
management of peripheral vestibular disease and avoiding
dangerous misdiagnoses of central ones. Patients with the INTRODUCTION
acute vestibular syndrome mostly have vestibular neuritis,
but some have stroke. Data suggest that focused eye move- Patients with dizziness account for just over 3% of all
ment examinations, at least when performed by specialists, emergency department (ED) visits (1). Because of the
are more sensitive for detecting early stroke than brain im- wide differential diagnosis, these individuals can be diffi-
aging, including diffusion-weighted magnetic resonance im- cult to diagnose. Although most have benign vestibular
aging. Patients with the triggered episodic vestibular problems, or general medical causes that are often
syndrome mostly have benign paroxysmal positional vertigo
apparent from the clinical context, co-symptoms, and vi-
tal signs, some have posterior circulation stroke and other
Streaming video: One brief real-time video clip that accom- serious central nervous system (CNS) causes. The
panies this article is available in streaming video at www.jour traditional diagnostic paradigm is based on symptom
nals.elsevierhealth.com/periodicals/jem. Click on Video Clips quality—‘‘what do you mean dizzy?’’ According to this
1, 2, and 3. system, a patient with ‘‘vertigo’’ has a different list of

RECEIVED: 3 June 2015; FINAL SUBMISSION RECEIVED: 6 October 2015;

ACCEPTED: 13 October 2015

617
618 J. A. Edlow and D. Newman-Toker

possible causes than those who endorse ‘‘lightheaded- and epidemiological context. The word used by the pa-
ness’’ (2). tient (e.g., vertigo vs. lightheadedness, or other terms)
During the last decade, mounting research suggests is unimportant, just as the patient descriptor of what
that this paradigm, based on data published more than kind of headache or chest pain (e.g., pressure vs. burning
40 years ago, is faulty and that the specific word that vs. stabbing) has very limited differential diagnostic util-
the patient endorses is not diagnostically meaningful ity. In one study, dizzy ED patients changed their type of
(3,4). A new paradigm based on the timing and triggers dizziness 50% of the time when asked repeated ques-
of dizziness has been proposed (4–7). According to this tions in a different sequence within 10 min (9). In this
new paradigm, patients are grouped into major same study, patients were far more consistent with their
categories (Table 1) based on the timing and trigger reporting of timing and triggers of symptoms. For
pattern of their dizziness symptoms (8). Understanding simplicity, in this article we will use the term dizziness
these three clinical groups is key to the use of the physical to mean any vestibular symptoms, including vertigo.
examination. One of the most serious misdiagnoses that can occur in
Patients with the acute vestibular syndrome have patients with dizziness is missed posterior circulation
monophasic acute dizziness that begins abruptly or very stroke. Because many symptoms of vertebrobasilar
rapidly and persists continuously for days. Although ischemia, especially in the cerebellum, are nonspecific
they will usually feel less symptomatic when not moving, (e.g., headache, dizziness, vomiting), misdiagnosis is
they are still dizzy at rest. The critical distinction is that not uncommon (10). In one study of 1,666 ED patients
their dizziness may be exacerbated (i.e., from an presenting with acute dizziness, one-third of the patients
abnormal baseline) by movement but it is not triggered who had strokes were missed in the ED (11). Numerous
(i.e., from a normal baseline) by movement. Patients studies using various methodologies have identified a
with the triggered episodic vestibular syndrome (t-EVS) presentation of dizziness and posterior circulation loca-
have repetitive episodes of dizziness that are triggered tion as being risk factors for missed stroke diagnosis
by some event. They are completely asymptomatic at (12–17). Misdiagnosis can result in poor patient
rest and will develop dizziness (usually lasting < 1 min) outcomes (18).
that is reliably triggered by a specific head movement Misdiagnosis is partly due to physicians’ misconcep-
or postural shift (i.e., standing or sitting up). Finally, pa- tions about factors that allow the diagnosis of a peripheral
tients with the spontaneous episodic vestibular syndrome process (3,19). Common examples include the false
(s-EVS) have multiple episodes of dizziness (usually last- notion that symptoms that worsen with movement
ing many minutes to many hours) that comes on without confirm a peripheral vestibular process and that a
any clear identifiable trigger, even if there appear to be normal brain computed tomography (CT) scan is
contextual predisposing factors (e.g., sleep deprivation, reassuring in excluding ischemic stroke (3,19). A recent
stress, or hormonal changes). Between episodes, they analysis of patients who presented to Canadian EDs
are completely asymptomatic. Table 1 shows the more with dizziness and were subsequently discharged
common benign and serious causes for each of these three showed that patients who had a (negative) CT scan had
vestibular syndromes. a twofold increased likelihood of stroke compared with
Accordingly, taking a history from patients with dizzi- those who were not scanned, suggesting that the
ness should be exactly like taking a history from patients emergency physicians (EPs) were correctly considering
with headache or chest pain and focuses on the duration, stroke but were falsely reassured by a negative CT
episodic or constant nature of the symptom, triggers if it result (20). An additional finding was that CT scans
is episodic, and other factors and associated symptoms were being overused and not in accordance with

Table 1. Timing-and-Trigger-Based Vestibular Syndromes in Acute Dizziness

Vestibular* Syndrome Common Benign Cause Dangerous Cause(s)

Triggered† episodic vestibular syndrome BPPV Posterior fossa tumor

Spontaneous episodic vestibular syndrome Vestibular migraine TIA
Cardiac dysrhythmia
Acute vestibular syndrome Vestibular neuritis Stroke

BPPV = benign paroxysmal positional vertigo; TIA = transient ischemic attack.

* The term vestibular here connotes vestibular symptoms (e.g., dizziness or vertigo or imbalance or lightheadedness), rather than under-
lying vestibular causes (e.g., benign paroxysmal positional vertigo, vestibular neuritis).
† Dizziness is ‘‘triggered’’ when it is brought on from a baseline of no symptoms, as in positional vertigo due to BPPV. This is distinguished
from dizziness that is ‘‘exacerbated’’ from a milder baseline state, such exacerbations are common in acute vestibular syndrome (AVS),
whether peripheral (neuritis) or central (stroke).
Using Physical Examination to Diagnose Dizziness 619

recommendations. The sensitivity of brain CT for poste- stroke (41). The next most common is multiple sclerosis,
rior circulation acute ischemic stroke is low (7%–42%) which, although an important CNS condition, is not
(21–24). immediately life threatening; serious harm is unlikely
Finally, although CT is an excellent test for an intrace- to occur in a patient who is diagnosed later in follow-
rebral hemorrhage, an intracerebral hemorrhage that up. Less common causes of AVS (medical and other
mimics a benign dizziness presentation (no obvious, neurological conditions) will be suggested by the clinical
worrisome physical signs such as lethargy or hemipare- context, vital signs, and basic laboratory testing (42).
sis) is rare (25). The total annual cost for evaluating dizzy Therefore, we will focus on the major differential diag-
patients in United States (US) EDs now exceeds $4 nosis here—vestibular neuritis vs. stroke.
billion (26). Imaging accounts for 12% of the ED costs, Recent literature has highlighted the importance of
75% of which is attributable to CT scans. Most of these bedside testing in these patients. However, an important
CT scans are not clinically useful and, as mentioned, first principle in dizziness is that a general neurological
can be falsely reassuring (20,27–29). physical examination needs to be done. Cranial nerve,
Fortunately, growing evidence suggests that the phys- cerebellar, and gait testing are particularly important. In
ical examination can help physicians confidently make a patients with posterior circulation strokes, approximately
specific peripheral vestibular diagnosis on the one hand, 30%–60% will have one of these findings (10). Important
or suggest stroke on the other. Most of the data come parts of the examination include testing finger-to-nose (or
from bedside examinations performed by specialists finger-to-chin, to avoid patients poking themselves in the
(stroke neurologists) or subspecialists (neuro-otologists) eye) and heel-to-shin, facial motor and sensory symme-
(30–32). However, there is nothing ‘‘magical’’ about try, dysarthria, and gait testing. Ideally, have the patient
these clinical tests and EPs can learn to perform and walk unassisted, but for patients too symptomatic to
interpret them (33). walk, test truncal ataxia by asking the patient to sit up-
Modern medical practice increasingly favors testing right in the stretcher without holding onto the side rails.
over physical examination. Magnetic resonance imaging Lateral medullary stroke, an important cause of AVS,
(MRI), which is becoming increasingly available in real merits special attention. These patients often complain of
time, has been suggested as the gold standard for diag- dysarthria, dysphagia, or hoarseness due to lower cranial
nosing acute ischemic stroke (34). However, it is neuropathy (43). These patients may have Horner’s syn-
important to recognize that early MRI, even with drome, with subtle ptosis and anisocoria, which may
diffusion-weighted imaging sequences (DWI), has impor- only be evident in dim light. The most common physical
tant limitations in diagnosing acute ischemic stroke. examination finding is hemifacial decreased pain and
Studies using delayed DWI-MRI or final clinical diag- temperature sensation. Routine testing of light touch
nosis as the gold standard have found frequent false neg- will miss this finding.
atives in the first 48 h after onset of symptoms. This is In patients with normal general neurological examina-
especially true for posterior fossa events (21,31,32,35– tions, specific bedside eye movement testing is very use-
38). It is also specifically true in dizzy patients (31,32,38). ful. In 2009, the HINTS study was published that assessed
Taken together, these data suggest that physicians are the value of neuro-ophthalmologic testing by a neuro-
using a diagnostic paradigm that is not consistent with otologist in patients presenting with an AVS and at least
current evidence, brain imaging, particularly CT, is over- one stroke risk factor. These investigators found that the
used, and misdiagnosis could be reduced. Fortunately, the physical examination identified all of the stroke patients,
physical examination is highly sensitive in these patients. as compared to DWI-MRI performed within 48 h of
This article will focus on how to use bedside physical ex- symptom onset, which was falsely negative in 12% of
amination to more accurately diagnose patients who pre- cases relative to a delayed MRI (31). The three tests in
sent with acute dizziness, vertigo, or other similar the HINTS battery are assessing the horizontal head im-
vestibular symptoms (Figure 1). pulse test (HIT) and testing for nystagmus and skew devi-
ation. The findings expected in peripheral vs. central
disorders are shown in Table 2.
DISCUSSION
Normally, the first step is testing for nystagmus.
Patients Who Present with Acute Vestibular Syndrome Studies suggest that EP documentation of nystagmus is
both arbitrary and not detailed enough to be diagnosti-
Acute vestibular syndrome (AVS) is acute, rapid-onset, cally useful. In a chart review of 1,485 dizzy patients
persistent dizziness with a monophasic course that re- from various EDs (some of which had charts that included
solves over days to weeks. The term vestibular does not a templated entry for nystagmus), EP reporting of
refer to a vestibular cause. The most common causes of nystagmus rose from 12% (in charts without the tem-
the AVS are vestibular neuritis and posterior circulation plate) to 95% (in those with charts with it) (44). In a
620 J. A. Edlow and D. Newman-Toker

Dizziness is the sensation of disturbed or impaired spatial orientation without a false or

distorted sense of motion. This includes sensations sometimes referred to as giddiness,

lightheadedness, or non-speciϔic dizziness, but does not include vertigo.

Vertigo is the sensation of self-motion (of head/body) when no self-motion is occurring or

the sensation of distorted self-motion during an otherwise normal head movement. This

includes a false sense of rotation (e.g., spinning, rocking), linear motion (e.g., feeling of

falling downward, as in an elevator), or static tilt relative to gravity.

Presyncope (also near syncope or faintness) is the sensation of impending loss of

consciousness. This sensation may or may not be followed by syncope.

Syncope (also faint) is transient loss of consciousness due to transient global cerebral

hypoperfusion characterized by rapid onset, short duration, and spontaneous complete

recovery. Syncope usually leads to loss of postural control and falling.

Unsteadiness (formerly known as disequilibrium) is the feeling of being unstable while

seated, standing, or walking without a particular directional preference.

Directional pulsion is the feeling of being unstable with a tendency to veer or fall in a

particular direction while seated, standing, or walking.

Figure 1. International expert consensus definitions related to dizziness and vertigo (39,40).

related study, 887 of 1,091 ED patients with dizziness had cant knowledge gap regarding nystagmus testing and inter-
some documentation of testing for nystagmus (which was pretation. Understanding that nystagmus testing can
present in 185 of the 887) (45). Of these 185 patients with improve the accurate diagnosis and safe disposition in
nystagmus, sufficient information (beyond merely ‘‘pre- ED patients is an important prerequisite to closing this gap.
sent’’) to allow a meaningful inference about the cause Just as with any test, knowing what one is looking for
was recorded in only 10 cases (5%). In 140 cases and careful observation are key. There are different forms
receiving a final ED diagnosis of benign paroxysmal po- of nystagmus, the most common of which is jerk
sitional vertigo (BPPV), vestibular neuritis, or labyrinthi- nystagmus, in which the eyes slowly drift off target
tis, the EP’s description of the nystagmus was (this is actually the pathologic phase) and then snap
inconsistent with the final diagnosis in 81% of cases. back to the target (by convention, this ‘‘fast phase’’ is
In a smaller study of 28 ED patients with acute vertigo, the direction that is named). For example, if the eyes drift
EPs did not record any information about nystagmus in 9 rightward, then the fast component beats toward the pa-
patients (46). Taken together, these data suggest a signifi- tient’s left, this is called ‘‘left-beating’’ nystagmus.
Using Physical Examination to Diagnose Dizziness 621

Table 2. Eye Findings in the Acute Vestibular Syndrome Without New Hearing Loss

HINTS Eye Finding in AVS Peripheral Central

Head impulse Unilaterally abnormal VOR toward the Usually bilaterally normal VOR*
affected side
Nystagmus (straight ahead, Dominantly horizontal, direction-fixed, Often mimics peripheral; if direction-
gaze right/left) beating away from the affected side changing horizontal or dominantly
vertical or torsional, then central†
Test of skew (alternate cover test) Normal vertical eye alignment Often mimics peripheral; if skew
(i.e., no skew deviation) deviation is present, then central‡

AICA = anterior inferior cerebellar artery; AVS = acute vestibular syndrome; VOR = vestibulo-ocular reflex.
* Strokes in the AICA territory may produce a unilaterally abnormal VOR that mimics vestibular neuritis, but hearing loss is usually present
as a clue. If a patient has bilaterally abnormal VORs, this is also suspicious for a central lesion if nystagmus is present (AICA stroke or Wer-
nicke’s syndrome).
† Inferior branch vestibular neuritis will present with downbeat-torsional nystagmus, but this is a fairly rare disorder. This form of
nystagmus in an AVS patient should prompt a stroke work-up.
‡ Skew deviation evident by bedside alternate cover testing does occur in 2% of peripheral vestibular cases, but is much more common
with central lesions. Presence of skew increases the odds of stroke or other central cause 20-fold in the AVS, so its presence mandates a
stroke work-up (41).

For dominantly horizontal nystagmus, descriptors are rightward gaze and left-beating on leftward gaze), this is
right-beating or left-beating; for vertical, up-beating or called ‘‘direction-changing’’ nystagmus (almost always
down-beating. Torsional movements are a little trickier indicating a central process). The change in the direction
to name. It is best to avoid the terms clockwise and coun- of the fast phase is the important finding.
terclockwise, because the intuitive approach is to ‘‘call it A key point is that it is not merely the presence (or
as you see it’’ (i.e., from the examiner’s perspective), but absence) of nystagmus that is diagnostic, but rather the
the official correct terminology for all eye movements is characteristics of the nystagmus (e.g., vector, direction,
from the patient’s perspective. So the preferred approach change in different gaze positions, or with different
is instead to look at movement of the 12 o’clock position viewing conditions). All patients with vestibular neuritis
on the eyeball, and call it ‘‘right-ear beating’’ or ‘‘left-ear should have nystagmus if one examines carefully in the
beating,’’ which is unambiguous and the same from the first few days of the illness. However, the vast majority
patient or physician perspective. of patients with cerebellar stroke who present with dizzi-
In practice, first examine for spontaneous nystagmus, ness or vertigo also have nystagmus (47). Understanding
and then test for gaze-evoked nystagmus. For sponta- this point also helps to close the knowledge gap
neous nystagmus, carefully observe the patient’s eyes mentioned here.
while they are looking straight ahead in neutral (primary) Several additional caveats in testing for nystagmus in
gaze. Observe for jerk-like deflections and note the direc- AVS are important:
tion in which the fast phase beats and if the primary
movement vector is horizontal, vertical, or torsional. In  Patients may have physiological end-gaze
the majority of dizzy patients, the nystagmus will look nystagmus (this finding, of course, could be seen
about the same in both eyes, unless one of them is para- in a patient with some other medical cause of dizzi-
lyzed (e.g., from a brainstem stroke), so you can usually ness)
come closer to the patient, focusing your attention on just  Direction-changing nystagmus indicates a CNS
one eye when determining the direction. This is particu- process, but not necessarily a serious one (e.g.,
larly important if the nystagmus is mild. phenytoin or alcohol intoxication)
Next, test for gaze-evoked nystagmus by asking the  Occasional patients with inferior branch vestibular
patient to follow one’s finger or another object (e.g., a neuritis will show dominantly torsional and vertical
penlight) placed about 25 cm away from the patient’s nystagmus, mimicking CNS disease
face. Move the object to the right and left and up and  AVS patients with nystagmus at rest do not require
down. Encourage the patient to keep their eyes wide positional testing and pursuing such testing will un-
open. Note the presence of nystagmus in each direction necessarily aggravate patient symptoms
of gaze and note the direction of its rapid phase. If the di-  As a general rule, visual fixation (i.e., having the pa-
rection of the rapid phase is the same (e.g., right-beating) tient stare at a fixed target) will reduce, and some-
no matter if the patient is looking to the right, left, up, or times completely extinguish, the spontaneous
down, this is called ‘‘direction-fixed’’ nystagmus, more nystagmus of vestibular neuritis; this is also seen
often suggesting a peripheral process, while if the direc- with some acute posterior fossa strokes (48). There-
tion of the fast component changes (e.g., right-beating on fore, lack of reduction of nystagmus with visual
622 J. A. Edlow and D. Newman-Toker

fixation suggests a central cause, but reduced cates a peripheral process, usually vestibular neuritis.
nystagmus with visual fixation is uninformative. The absence of a corrective saccade in AVS is consistent
with a stroke. Note, it is somewhat counterintuitive that a
Some patients have physiological end-gaze normal bedside finding predicts a dangerous disease. Pa-
nystagmus. When these patients look to the extreme right tients with cerebellar stroke have a negative (normal) HIT
or left lateral gaze they will have a few beats of symmet- because the VOR does not loop through the cerebellum
ric, rapidly extinguishing nystagmus that beats to the (47,54).
right on rightward gaze and to the left on leftward gaze, On the other hand, occasional patients with posterior
mimicking a central lesion. Therefore, if a few beats of circulation stroke will have a falsely positive (i.e.,
gaze-evoked nystagmus occur at end-gaze, one should abnormal) HIT, usually from a lateral brainstem infarct
also test the patient with the target about 45 off the where the vestibular nerve enters the brainstem (due to
midline to see if the nystagmus is still present. Physicians an anterior inferior cerebellar artery [AICA] stroke) or
who routinely test for nystagmus during lateral gaze in from an infarction directly involving the inner ear (laby-
patients who are not dizzy will quickly develop a good rinthine stroke). Adding a bedside test of hearing
sense of what physiological nystagmus looks like. (‘‘HINTS plus’’) will help to pick up the occasional
Remember that ‘‘central’’ does not always mean a AICA stroke because hearing is usually involved in this
stroke or posterior fossa structural lesion. For example, lesion (55).
thiamine deficiency is a dangerous ‘‘central’’ problem, The fact that it is the ‘‘positive’’ test that is generally
but one that needs urgent medical, rather than neurolog- reassuring (because it indicates a peripheral process) is
ical, treatment. Many patients with a therapeutic a key point. Usually in medicine, it is the ‘‘positive’’
phenytoin level or acute alcohol intoxication will have test that indicates a serious diagnosis in question; here,
direction-changing nystagmus from direct effects on the it is the opposite. This means that one should only use
brainstem and cerebellum. the HIT in patients with an AVS presentation. Otherwise,
The vestibular nerve has a superior and inferior branch. the HIT will be ‘‘negative’’ (i.e., suggest a stroke). If one
Most patients with vestibular neuritis have superior branch performs the HIT on a patient with dizziness due to dehy-
involvement, which generally affects the horizontal and dration or pulmonary embolism (or on a healthy subject),
anterior canal on one side (49 51). Clinically, this means the test will be ‘‘negative’’ (i.e., normal and falsely sug-
that the nystagmus, although predominantly horizontal, gestive of stroke). This means case selection is critical,
will usually have a slight torsional and vertical and it is generally unwise to rely on a normal HIT result
component as well. In the unusual case of inferior branch if the patient has neither spontaneous nor gaze-evoked
vestibular neuritis, patients will exhibit mixed torsional nystagmus, because these are key components of the
and down-beating nystagmus, which mimics a central AVS presentation.
cause (52). Assuming that all AVS patients with this form The third component of HINTS battery is testing for
of nystagmus have a central process will result in referring skew deviation (a vertical misalignment of the eyes due
or working up these few patients as if they have had a to imbalance in the gravity-sensing pathways) that is
stroke, but this ‘‘error’’ favors patient safety. not very sensitive (30%), but is very specific (98%) for
In a patient with AVS and nystagmus sitting or lying a brainstem lesion (41). To test for skew deviation, the
still in the bed without head movement, positional ma- examiner uses the ‘‘alternate cover’’ test. With the patient
neuvers, such as the Dix-Hallpike test, are not helpful looking directly at the examiner’s nose, the physician
and will cause the patient unnecessary discomfort. This alternately covers the right eye, then the left eye, and con-
underscores the notion that different physical examina- tinues alternating back and forth. In patients with skew
tion components are useful for different clinical syn- deviation, each time the covered eye is uncovered, there
dromes (4). will be a slight vertical correction (on one side you will
The next component of the HINTS battery is the HIT, a see an upward correction and on the other you will see
test of the vestibulo-ocular reflex (VOR) described in a downward correction). It is not important which eye
1988 (53). The test should be performed several times, the examiner focuses on first, but it is important to focus
very quickly toward each ear using a lateral to center mo- on one eye at a time over several cycles of alternate
tion, over a very short distance (no more than 20 ) covering, rather than following the ‘‘cover’’ from side
(HINTS Video 1; available at http://content.lib.utah. to side, to be able to see the corrective eye movements.
edu/cdm/ref/collection/ehsl-dent/id/6). This lateral-to- A normal response is no vertical correction, and an
midline motion reduces the possibility of over- abnormal response is suspicious for stroke in a patient
torqueing the head and may be safer. The presence of a presenting with AVS.
corrective saccade (fast eye movement) is a ‘‘positive’’ There is no published evidence regarding the ideal or-
test and a positive test (abnormal VOR) generally indi- der of these tests and it is likely that the order does not
Using Physical Examination to Diagnose Dizziness 623

matter much. The single most potent test is the HIT (41). with a complaint of postural (orthostatic) lightheadedness
Nevertheless, we prefer testing nystagmus first because (56). It is therefore important to ask those with orthostatic
this is the least intrusive to the patient and because it helps symptoms whether they also have symptoms on lying
frame the problem—if nystagmus is absent, one must back or rolling in bed, neither typical of orthostatic hypo-
question the syndromic classification as AVS, so the tension (Figure 2B).
interpretation of the HIT result might be altered. Some BPPV is common (57). Patients usually describe brief
test for skew deviation second, because it is also less episodes of dizziness that are triggered by head move-
intrusive, but others place it last because it is least likely ment. The common trigger of rolling over in bed can
to add to the diagnosis. lead to patients being awakened from sleep by dizziness.
The HINTS battery of testing has primarily been stud- The pathophysiology is well described in reviews and
ied in the hands of specialists (30–32). Although HINTS specialty guidelines (58 60). In the usual case, calcium
appears to be accurate independent of the presence or carbonate otoliths from the utricle or saccule (structures
absence of other neurological signs, implicit in the in the inner ear that sense linear motion and tilting)
examinations done in these studies is the targeted come loose and lodge in one of the three paired
neurological examination, especially gait testing. semicircular canals that sense rotational motion of the
Therefore, it may be prudent to pose five questions in head. This state is called canalolithiasis (literally,
patients who present with the AVS (Figure 2A). ‘‘rocks in the canals’’). During normal head movements,
these displaced otoliths slide inside the affected canal.
1. Is the patient unable to sit or stand without assis-
This nonphysiologic stimulus is conducted via the
tance?
ipsilateral vestibular nerve to the brain, where it is
2. Does the patient have any new finding on a targeted
interpreted as head motion, even though no real motion
general neurological examination?
is occurring (vertigo), or as a more vague sense of
3. Does the patient have worrisome spontaneous or
spatial disorientation (dizziness). Patients with BPPV
gaze-evoked nystagmus (predominantly vertical
have normal hearing and do not have other neurological
or torsional, or direction-changing horizontal)?
findings.
4. Does the patient have a negative HIT (no corrective
Delays in diagnosis of BPPV are common (61). Pa-
saccade on either side)?
tients who describe ‘‘vertigo’’ are diagnosed sooner
5. Does the patient have skew deviation (vertical eye
than those who endorse ‘‘dizziness’’ or ‘‘lightheaded-
misalignment)?
ness’’ (56). The classic history is one of episodes of dizzi-
If the answer to any of these five questions is ‘‘yes,’’ ness that are abruptly triggered by head movement and
then the patient needs an evaluation for stroke. If the resolve in less than a minute upon stopping head motion
answer to all five questions is ‘‘no,’’ then the patient (even while maintaining the head in the offending trigger
almost certainly has a peripheral cause for the AVS. position). Some patients overestimate the duration
This approach acknowledges that none of the five tests because they trigger back-to-back episodes, have
can be used individually to exclude stroke because none lingering mild symptoms of nausea or unsteadiness, or
has 100% sensitivity. As described here, CT is inadequate fear of triggering another episode. Other patients may
to diagnose ischemic stroke in the posterior fossa. Like- initially endorse an episode lasting days, but one can
wise, because it is clear that early MRI is falsely negative easily tease out that they are describing the total illness
in 10% to 20% of these strokes, MRI is also not ideal for duration, during which brief, triggered episodes have
deciding whether or not to admit for stroke work-up been occurring. Many patients undergo extensive, unnec-
(31,32,55). Put another way, physical examination is essary work-ups, including MRI, before the correct diag-
superior to DWI-MRI in the early diagnosis of posterior nosis and treatment (62).
circulation stroke in patients with AVS, and even a nega- There are different diagnostic and physical maneuvers
tive DWI-MRI is no guarantee the patient does not have for each semicircular canal (Table 3) (58,59). The
stroke. posterior canal is the most frequently affected due to its
anatomy (a dependent loop of the canal lies beneath the
Triggered Episodic Vestibular Syndrome canal’s opening when we are standing or sitting
upright). Therefore, this canal is generally tested first
Patients with the t-EVS are the second group in whom with the Dix-Hallpike maneuver (Video 2; available at:
physical examination has an especially important impact. http://content.lib.utah.edu/cdm/ref/collection/ehsl-dent/
The most common causes are BPPVand orthostatic hypo- id/3). Each side is tested independently. The patient
tension. Because various conditions causing orthostatic should be asymptomatic at the start of the test. In typical
hypotension are well known to EPs, we will not discuss cases, symptoms and nystagmus should begin after a few
these further except to note that BPPV may present seconds of latency and resolve in < 30 s. The expected
624 J. A. Edlow and D. Newman-Toker

Figure 2. (A) Five necessary ‘‘safe-to-go’’ findings in the acute vestibular syndrome (AVS). Together, these findings reduce
the odds of stroke by at least 50-fold (41). This means that a typical AVS patient with a 25% pretest probability of stroke
has < 1% post-test probability of stroke if all five features are present. By contrast, a negative diffusion-weighted imaging
sequences magnetic resonance imaging obtained in the first 24 h after onset of AVS symptoms would reduce the odds of
stroke by about fivefold, giving a post-test probability of roughly 7%. (B) Bedside diagnosis for patients with triggered
episodic vestibular syndrome (t-EVS). By definition, patients with t-EVS are 100% asymptomatic at rest, and develop dizzi-
ness that is triggered by movement only. BPPV = benign paroxysmal positional vertigo.
Using Physical Examination to Diagnose Dizziness 625

Table 3. Positional Nystagmus Findings in Triggered, Episodic Vestibular Syndrome

Positional Tests in t-EVS BPPV (Posterior) BPPV (Horizontal) Central

Dix-Hallpike Upbeat-torsional* None† Variable direction (downbeat or

5–30 s horizontal; almost never upbeat)
No spontaneous reversal Variable duration (often > 90 s)
No spontaneous reversal
Supine roll test None† Pure horizontal‡ Variable direction (downbeat or
30–90 s horizontal; almost never upbeat)
Spontaneous reversal typical Variable duration (often > 90 s)
No spontaneous reversal

BPPV = benign paroxysmal positional vertigo; t-EVS = triggered, episodic vestibular syndrome.
* The nystagmus of posterior canal BPPV will have a prominent torsional component, and the 12 o’clock pole of the eye will beat toward
the down-facing (tested) ear. Although the nystagmus will reverse on arising from the Dix-Hallpike position, there will be no spontaneous
reversal.
† Although the Dix-Hallpike test is fairly specific to posterior canal BPPV and the supine roll test to horizontal canal BPPV, the maneuvers
may sometimes stimulate the other canal. If so, the nystagmus direction will depend on the affected canal, not on the type of maneuver
eliciting the nystagmus. The nystagmus may be considerably weaker and less evident than when using the ‘‘correct’’ maneuver.
‡ The nystagmus of horizontal canal BPPV may beat toward the down ear or away from it. The nystagmus will often crescendo, then slow
down and reverse spontaneously even without moving the head. When the opposite side is tested, the nystagmus will usually beat in the
opposite direction (e.g., if right-beating initially with right ear down, then left-beating initially with left ear down).

nystagmus should be mixed upbeat torsional. The posi- usually due to brainstem stroke. Thus, it is essential to
tive side identifies the affected posterior canal (right ear separate the clinical syndromes not only to target the ex-
down = right posterior canal), which, in turn, guides the amination, but to correctly interpret its results.
direction of the Epley maneuver (58). EPs can diagnose If patients whose histories suggest BPPV have nega-
and treat BPPV with a curative canalith repositioning pro- tive Dix-Hallpike tests (or dominantly horizontal
cedure (CRP) (63). nystagmus), then specifically testing the horizontal (also
The Dix-Hallpike test also simultaneously examines referred to as the lateral) canal is indicated using the su-
the anterior canal; however, this canal is rarely involved pine head roll test (Video 3; available at http://content.
and the characteristic nystagmus is down-beating, which lib.utah.edu/cdm/ref/collection/ehsl-dent/id/12) (58). Pa-
is a typical nystagmus for central mimics (e.g., posterior tients with horizontal canal BPPV will test positive no
fossa tumors). For both of these reasons, it is probably matter to which side the head is turned, but one side will
safer to consult a specialist for further testing to exclude be more symptomatic and the nystagmus will be more
a CNS condition rather than attempt to diagnose anterior intense. Nystagmus will beat horizontally (different than
canal BPPV (60). the upbeat-torsional nystagmus of posterior canal
It is a common misconception that dizziness made BPPV) and will generally be more severe and prolonged
worse by head motion (including positional testing) is a than the posterior canal variant (about 60 s). The side pro-
sign of peripheral vestibular disease, and, specifically, voking more intense nystagmus is typically the affected
BPPV (3). We must draw a critical distinction between side that will guide the repositioning maneuver (58).
dizziness that is exacerbated by head motion and dizzi- Up to 25% of patients with a classic t-EVS history for
ness that is triggered by head motion. An AVS patient BPPV will not have obvious nystagmus on observation
(symptomatic at baseline) examined using the Dix- during the Dix-Hallpike test (60,64,65). This may be
Hallpike test will almost invariably have increased symp- from a mild case, partial spontaneous resolution, or
toms and increased intensity nystagmus (exacerbated). suboptimal test conditions (e.g., lack of special goggles
These patients will feel worse with any head movement. to block visual fixation). Regardless, these patients can
However, if the Dix-Hallpike test induces symptoms and be treated successfully in the standard fashion using the
nystagmus from a t-EVS patient with a history of brief, symptomatic side as a guide to determine the correct
recurrent spells triggered by specific head movements treatment maneuver. Importantly, however, there is no
(no symptoms at baseline), this is generally diagnostic characteristic nystagmus to support a specific diagnosis,
of BPPV. This is why the Dix-Hallpike test should be treatment failures (i.e., lack of immediate and
reserved for use in patients with t-EVS. significant relief on completing the maneuver) should
Furthermore, the same nystagmus has opposite mean- immediately prompt a search for other causes.
ing in these two distinct vestibular syndromes. In t-EVS, a Although uncommon, CPPV can mimic BPPV. This is
burst of upbeat-torsional nystagmus triggered by the Dix- usually caused by demyelinating lesions, strokes, or tu-
Hallpike position means BPPV. In the AVS, upbeat mors in the posterior fossa (66). Clues to CPPV include
torsional nystagmus (the same finding) is a central sign, the presence of other neurological symptoms or signs
626 J. A. Edlow and D. Newman-Toker

that do not occur with BPPV, presence of nystagmus on not difficult. Unfortunately, this is often not the case.
positional testing but without associated dizziness, Because migraine is a central phenomenon, some of these
down-beating nystagmus, lack of latency to onset of patients may have physical findings that are suggestive of
nystagmus, persistence of nystagmus for >90 s, and fail- a CNS process (if they are symptomatic at the time of pre-
ure of symptoms or nystagmus to resolve with a CRP sentation). The type the nystagmus is variable and could
(66). be horizontal (direction-fixed or direction-changing), ver-
Uncommon variant forms of BPPV include multiple- tical, or torsional (73).
canal disease or otoliths that stick to the cupula In patients without a well-established history of
(cupulolithiasis) rather than float freely in the canal (ca- vestibular migraine who are presenting with persistent
nalolithiasis). These disorders can produce complex pat- symptoms and physical findings consistent with a central
terns of nystagmus that defy simple rules and may mimic problem, it is safest to proceed as if the patient has an
CNS disease. Another atypical nystagmus variant is AVS with central features.
pseudo-spontaneous nystagmus, seen only in patients
with horizontal canal BPPV; this disappears, depending CONCLUSIONS
on the attitude of the head and neck in the anterior-
posterior plane (67). Each of these can lead to confusing ED patients with a chief symptom of new dizziness fall
findings for the nonspecialist. As a general principle, in a into three major categories (AVS, t-EVS, s-EVS) based
patient whose history suggests BPPV, if there are atypical on their description of the timing, triggers, context of
examination features or if a physician cannot elicit typical the dizziness, and other associated symptoms. Studies
findings that resolve after the appropriate CRP, specialist of current ED practice reveals knowledge gaps in bedside
consultation is probably the safest course of action. diagnosis of these patients, resulting in misdiagnosis of
serious causes such as stroke, unnecessary use of neuro-
Spontaneous Episodic Vestibular Syndrome imaging, and failure to institute specific treatment for
many patients with inner ear causes of dizziness. Educa-
Most patients with the s-EVS will be asymptomatic at the tional efforts targeted at improving EP physical examina-
time of presentation because the primary symptoms from tion skills have the potential to help EPs confidently make
an episode usually resolve within a few hours. Because specific, correct diagnoses. This, in turn, could lead to
the findings are, by definition, not triggered, physical ex- lower rates of misdiagnosis, decreased utilization of
amination is less useful in this situation because the ex- expensive imaging studies, and an increase in prompt,
amination is usually totally normal. This is akin to correct treatments, thereby improving patient outcomes.
patients with transient ischemic attack (TIA) and stroke.
If a patient with what ultimately turns out to be a TIA is
still having symptoms or signs at presentation, one should REFERENCES
proceed as if the patient is having a stroke, and search for
1. Newman-Toker DE, Hsieh YH, Camargo CA Jr, Pelletier AJ,
typical eye signs using HINTS, described here. Common Butchy GT, Edlow JA. Spectrum of dizziness visits to US emer-
causes of the s-EVS include vestibular migraine and Me- gency departments: cross-sectional analysis from a nationally repre-
nière’s disease, but TIA is the serious cause not to be sentative sample. Mayo Clin Proc 2008;83:765–75.
2. Drachman DA, Hart CW. An approach to the dizzy patient.
missed. Neurology 1972;22:323–34.
Historically, isolated dizziness was not considered to 3. Kerber KA, Newman-Toker DE. Misdiagnosing dizzy patients:
be a TIA (68). Recent data clearly show that this is not common pitfalls in clinical practice. Neurol Clin 2015;33:564–76.
4. Newman-Toker DE, Edlow JA. TiTrATE: a novel approach to diag-
true (69,70). In one study of 1,141 stroke patients, prior nosing acute dizziness and vertigo. Neurol Clin 2015;33:577–99.
transient brainstem symptoms were 15 times more 5. Edlow JA. Diagnosing dizziness: we are teaching the wrong para-
common in patients with posterior circulation events digm!. Acad Emerg Med 2013;20:1064–6.
6. Kerber KA. Vertigo and dizziness in the emergency department.
compared to those with anterior circulation strokes (70). Emerg Med Clin North Am 2009;27:39–50. viii.
Half of these events were isolated dizziness, and 9 of 7. Newman-Toker D. Diagnosing Dizziness in the Emergency Depart-
10 who presented for medical attention were initially mis- ment: Why ‘‘What Do You Mean by ‘Dizzy’?’’ Should Not Be the
First Question You Ask [doctoral dissertation]. Baltimore, MD:
diagnosed. As with all TIAs, diagnosis is by history alone Johns Hopkins School of Medicine; 2007.
and its recognition is an important means of secondary 8. Bisdorff A, Staab J, Newman-Toker D. Overview of the interna-
stroke prevention (71,72). tional classification of vestibular disorders. Neurol Clin 2015;33:
541–50.
The most common cause of s-EVS is vestibular 9. Newman-Toker DE, Cannon LM, Stofferahn ME, Rothman RE,
migraine. If the patient has had recurrent, stereotyped ep- Hsieh YH, Zee DS. Imprecision in patient reports of dizziness
isodes for years that are associated with typical migraine symptom quality: a cross-sectional study conducted in an acute
care setting. Mayo Clin Proc 2007;82:1329–40.
features (e.g., photophobia, visual aura) and predisposing 10. Edlow JA, Newman-Toker DE, Savitz SI. Diagnosis and initial man-
factors (e.g., menstrual periods), the diagnosis is usually agement of cerebellar infarction. Lancet Neurol 2008;7:951–64.
Using Physical Examination to Diagnose Dizziness 627

11. Kerber KA, Brown DL, Lisabeth LD, Smith MA, Morgenstern LB. 33. Vanni S, Nazerian P, Casati C, et al. Can emergency physicians
Stroke among patients with dizziness, vertigo, and imbalance in the accurately and reliably assess acute vertigo in the emergency
emergency department: a population-based study. Stroke 2006;37: department? Emerg Med Australas 2015;27:126–31.
2484–7. 34. Schellinger PD, Bryan RN, Caplan LR, et al. Evidence-based guide-
12. Casani AP, Dallan I, Cerchiai N, Lenzi R, Cosottini M, Sellari- line: the role of diffusion and perfusion MRI for the diagnosis of
Franceschini S. Cerebellar infarctions mimicking acute peripheral acute ischemic stroke: report of the Therapeutics and Technology
vertigo: how to avoid misdiagnosis? Otolaryngol Head Neck Surg Assessment Subcommittee of the American Academy of
2013;148:475–81. Neurology. Neurology 2010;75:177–85.
13. Gargano JW, Wehner S, Reeves MJ. Presenting symptoms and 35. Frey LC, Sung GY, Tanabe J. Early false-negative diffusion-
onset-to-arrival time in patients with acute stroke and transient weighted imaging in brainstem infarction. J Stroke Cerebrovasc
ischemic attack. J Stroke Cerebrovasc Dis 2011;20:494–502. Dis 2002;11:51–3.
14. Kim AS, Fullerton HJ, Johnston SC. Risk of vascular events in 36. Oppenheim C, Stanescu R, Dormont D, et al. False-negative
emergency department patients discharged home with diagnosis diffusion-weighted MR findings in acute ischemic stroke. AJNR
of dizziness or vertigo. Ann Emerg Med 2011;57:34–41. Am J Neuroradiol 2000;21:1434–40.
15. Nakajima M, Hirano T, Uchino M. Patients with acute stroke 37. Sylaja PN, Coutts SB, Krol A, Hill MD, Demchuk AM, Group VS.
admitted on the second visit. J Stroke Cerebrovasc Dis 2008;17: When to expect negative diffusion-weighted images in stroke and
382–7. transient ischemic attack. Stroke 2008;39:1898–900.
16. Newman-Toker DE, Moy E, Valente E, Coffey R, Hines AL. Missed 38. Park M, Kim K, Lee N, Jung H, Chae S. The usefulness of MRI for
diagnosis of stroke in the ED: a cross-sectional analysis of a large acute isolated vertigo patients in the emergency department. J Int
population based sample. Diagnosis 2014;2:29–40. Adv Otol 2014;10:162–6.
17. Pope JV, Edlow JA. Avoiding misdiagnosis in patients with neuro- 39. Bisdorff A, Von Brevern M, Lempert T, Newman-Toker DE. Clas-
logical emergencies. Emerg Med Int 2012;2012:949275. sification of vestibular symptoms: towards an international classifi-
18. Savitz SI, Caplan LR, Edlow JA. Pitfalls in the diagnosis of cere- cation of vestibular disorders. J Vestib Res 2009;19:1–13.
bellar infarction. Acad Emerg Med 2007;14:63–8. 40. Task Force for the Diagnosis and Management of Syncope; Euro-
19. Newman-Toker DE, Camargo CA Jr, Hsieh YH, Pelletier AJ, pean Society of Cardiology (ESC); European Heart Rhythm Asso-
Edlow JA. Disconnect between charted vestibular diagnoses and ciation (EHRA), et al. Guidelines for the diagnosis and management
emergency department management decisions: a cross-sectional of syncope (version 2009). Eur Heart J 2009;30:2631–71.
analysis from a nationally representative sample. Acad Emerg 41. Tarnutzer AA, Berkowitz AL, Robinson KA, Hsieh YH, Newman-
Med 2009;16:970–7. Toker DE. Acute vestibular syndrome: does my patient have a
20. Grewal K, Austin PC, Kapral MK, Lu H, Atzema CL. Missed stroke? A systematic and critical review of bedside diagnostic pre-
strokes using computed tomography imaging in patients with ver- dictors. Can Med Assoc J 2011;183:E571–92.
tigo: population-based cohort study. Stroke 2015;46:108–13. 42. Edlow JA, Newman-Toker DE. Medical and non-stroke neurolog-
21. Chalela JA, Kidwell CS, Nentwich LM, et al. Magnetic resonance ical causes of the acute vestibular syndrome. Neurol Clin 2015;
imaging and computed tomography in emergency assessment of pa- 33:699–716.
tients with suspected acute stroke: a prospective comparison. Lan- 43. Kim JS. Pure lateral medullary infarction: clinical-radiological cor-
cet 2007;369:293–8. relation of 130 acute, consecutive patients. Brain 2003;126:1864–
22. Hwang DY, Silva GS, Furie KL, Greer DM. Comparative sensitivity 72.
of computed tomography vs. magnetic resonance imaging for de- 44. Kerber KA, Hofer TP, Meurer WJ, Fendrick AM, Morgenstern LB.
tecting acute posterior fossa infarct. J Emerg Med 2012;42:559–65. Emergency department documentation templates: variability in
23. Kabra R, Robbie H, Connor SE. Diagnostic yield and impact of template selection and association with physical examination and
MRI for acute ischaemic stroke in patients presenting with dizziness test ordering in dizziness presentations. BMC Health Serv Res
and vertigo. Clin Radiol 2015;70:736–42. 2011;11:65.
24. Ozono Y, Kitahara T, Fukushima M, et al. Differential diagnosis of 45. Kerber KA, Morgenstern LB, Meurer WJ, et al. Nystagmus assess-
vertigo and dizziness in the emergency department. Acta Otolar- ments documented by emergency physicians in acute dizziness pre-
yngol 2014;134:140–5. sentations: a target for decision support? Acad Emerg Med 2011;18:
25. Kerber KA, Burke JF, Brown DL, et al. Does intracerebral haemor- 619–26.
rhage mimic benign dizziness presentations? A population based 46. Pavlin-Premrl D, Waterston J, McGuigan S, et al. Importance of
study. Emerg Med J 2012;29:43–6. spontaneous nystagmus detection in the differential diagnosis of
26. Saber Tehrani AS, Coughlan D, Hsieh YH, et al. Rising annual costs acute vertigo. J Clin Neurosci 2015;22:504–7.
of dizziness presentations to U.S. emergency departments. Acad 47. Lee H, Sohn SI, Cho YW, et al. Cerebellar infarction presenting iso-
Emerg Med 2013;20:689–96. lated vertigo: frequency and vascular topographical patterns.
27. Kerber KA, Schweigler L, West BT, Fendrick AM, Neurology 2006;67:1178–83.
Morgenstern LB. Value of computed tomography scans in ED dizzi- 48. Newman-Toker DE, Sharma P, Chowdhury M, Clemons TM, Zee DS,
ness visits: analysis from a nationally representative sample. Am J Della Santina CC. Penlight-cover test: a new bedside method to
Emerg Med 2010;28:1030–6. unmask nystagmus. J Neurol Neurosurg Psychiatry 2009;80:900–3.
28. Lawhn-Heath C, Buckle C, Christoforidis G, Straus C. Utility of 49. Jeong SH, Kim HJ, Kim JS. Vestibular neuritis. Semin Neurol 2013;
head CT in the evaluation of vertigo/dizziness in the emergency 33:185–94.
department. Emerg Radiol 2013;20:45–9. 50. Strupp M, Brandt T. Vestibular neuritis. Semin Neurol 2009;29:
29. Wasay M, Dubey N, Bakshi R. Dizziness and yield of emergency 509–19.
head CT scan: is it cost effective? Emerg Med J 2005;22:312. 51. Baloh RW. Clinical practice. Vestibular neuritis. N Engl J Med
30. Chen L, Lee W, Chambers BR, Dewey HM. Diagnostic accuracy of 2003;348:1027–32.
acute vestibular syndrome at the bedside in a stroke unit. J Neurol 52. Kim JS, Kim HJ. Inferior vestibular neuritis. J Neurol 2012;259:
2011;258:L855–61. 1553–60.
31. Kattah JC, Talkad AV, Wang DZ, Hsieh YH, Newman- 53. Halmagyi GM, Curthoys IS. A clinical sign of canal paresis. Arch
Toker DE. HINTS to diagnose stroke in the acute vestibular Neurol 1988;45:737–9.
syndrome: three-step bedside oculomotor examination more sen- 54. Newman-Toker DE, Kattah JC, Alvernia JE, Wang DZ. Normal
sitive than early MRI diffusion-weighted imaging. Stroke 2009; head impulse test differentiates acute cerebellar strokes from vestib-
40:3504–10. ular neuritis. Neurology 2008;70:2378–85.
32. Saber Tehrani AS, Kattah JC, Mantokoudis G, et al. Small strokes 55. Newman-Toker DE, Kerber KA, Hsieh YH, et al. HINTS outper-
causing severe vertigo: frequency of false-negative MRIs and non- forms ABCD2 to screen for stroke in acute continuous vertigo
lacunar mechanisms. Neurology 2014;83:169–73. and dizziness. Acad Emerg Med 2013;20:986–96.
628 J. A. Edlow and D. Newman-Toker

56. Lawson J, Johnson I, Bamiou DE, Newton JL. Benign paroxysmal 66. Soto-Varela A, Rossi-Izquierdo M, Sanchez-Sellero I, Santos-
positional vertigo: clinical characteristics of dizzy patients referred Perez S. Revised criteria for suspicion of non-benign positional ver-
to a Falls and Syncope Unit. QJM 2005;98:357–64. tigo. QJM 2013;106:317–21.
57. von Brevern M, Radtke A, Lezius F, et al. Epidemiology of benign 67. De Stefano A, Kulamarva G, Citraro L, Neri G, Croce A. Sponta-
paroxysmal positional vertigo: a population based study. J Neurol neous nystagmus in benign paroxysmal positional vertigo. Am J
Neurosurg Psychiatry 2007;78:710–5. Otolaryngol 2011;32:185–9.
58. Bhattacharyya N, Baugh RF, Orvidas L, et al. Clinical practice 68. The Ad Hoc Committee on the classification and outline of cerebro-
guideline: benign paroxysmal positional vertigo. Otolaryngol vascular disease II. Stroke 1975;6:566–616.
Head Neck Surg 2008;139(Suppl. 4):S47–81. 69. Hoshino T, Nagao T, Mizuno S, Shimizu S, Uchiyama S. Transient
59. Fife TD, Iverson DJ, Lempert T, et al. Practice parameter: therapies neurological attack before vertebrobasilar stroke. J Neurol Sci 2013;
for benign paroxysmal positional vertigo (an evidence-based re- 325:39–42.
view): report of the Quality Standards Subcommittee of the Amer- 70. Paul NL, Simoni M, Rothwell PM, Oxford Vascular S. Transient
ican Academy of Neurology. Neurology 2008;70:2067–74. isolated brainstem symptoms preceding posterior circulation stroke:
60. Kim JS, Zee DS. Clinical practice. Benign paroxysmal positional a population-based study. Lancet Neurol 2013;12:65–71.
vertigo. N Engl J Med 2014;370:1138–47. 71. Lavallee PC, Meseguer E, Abboud H, et al. A transient ischaemic
61. Fife D, FitzGerald JE. Do patients with benign paroxysmal posi- attack clinic with round-the-clock access (SOS-TIA): feasibility
tional vertigo receive prompt treatment? Analysis of waiting times and effects. Lancet Neurol 2007;6:953–60.
and human and financial costs associated with current practice. Int J 72. Rothwell PM, Giles MF, Chandratheva A, et al. Effect of urgent
Audiol 2005;44:50–7. treatment of transient ischaemic attack and minor stroke on early
62. Polensek SH, Tusa R. Unnecessary diagnostic tests often obtained recurrent stroke (EXPRESS study): a prospective population-
for benign paroxysmal positional vertigo. Med Sci Monit 2009; based sequential comparison. Lancet 2007;370:1432–42.
15:MT89–94. 73. Polensek SH, Tusa RJ. Nystagmus during attacks of vestibular
63. Chang AK, Schoeman G, Hill M. A randomized clinical trial to migraine: an aid in diagnosis. Audiol Neurootol 2010;15:
assess the efficacy of the Epley maneuver in the treatment of acute 241–6.
benign positional vertigo. Acad Emerg Med 2004;11:918–24.
64. Balatsouras DG, Korres SG. Subjective benign paroxysmal posi-
tional vertigo. Otolaryngol Head Neck Surg 2012;146:98–103. SUPPLEMENTARY DATA
65. Tan J, Yu D, Feng Y, et al. First-referral presentations of patients
with benign paroxysmal positional vertigo who were negative on
positional testing and who lacked nystagmus. Eur Arch Otorhinolar- Supplementary data related to this article can be found at
yngol 2015;272:3247–51. http://dx.doi.org/10.1016/j.jemermed.2015.10.040.

Streaming video: One brief real-time video clip that accompanies this article is
available in streaming video at www.journals.elsevierhealth.com/periodicals/jem.
Click on Video Clips 1, 2, and 3.