American Journal of Emergency Medicine (2012) 30, 587-591

Brief Report

ED patients with vertigo: can we identify clinical factors associated with acute stroke?^?

Maureen Chase MD ^a,?, Nina R. Joyce MPH ^a, Erin Carney BS ^a, Justin D. Salciccioli ^a,

Deborah Vinton MD ^b, Michael W. Donnino MD ^a, Jonathan A. Edlow MD ^a

^aBeth Israel Deaconess Medical Center, Department of Emergency Medicine, One Deaconess Road, West, Clinical Center 2,

Boston, MA 02215, USA

^bDenver Health System, Department of Emergency Medicine, Denver CO 80204, USA

Received 6 January 2011; revised 1 February 2011; accepted 2 February 2011

Abstract

Background: Vertigo is a common emergency department (ED) complaint with benign and serious etiologies with overlapping features. Misdiagnosis of acute stroke may result in significant morbidity and mortality. Magnetic resonance imaging (MRI) is superior to computer tomography (CT) for diagnosis of acute stroke but is costly with limited availability.

Objective: The aim of this study was to identify clinical characteristics associated with a cerebrovascular cause for vertigo.

Methods: We performed a retrospective chart review on patients with an MRI for vertigo, with or without additional historical or physical examination findings, over 18 months. Study patients were seen in the ED for vertigo within 2 weeks of MRI. Data collected included medical history, physical findings, and Imaging results. Fisher’s exact test was used to identify factors associated with the primary outcome, an acute stroke.

Results: There were 325 eligible patients; 131 were ED patients. Patients were 57 (+-18) years, and 53% were women. There were 12 ED patients with a new stroke (9.2%). Two variables were associated with acute stroke: a presenting complaint of gait instability (odds ratio, 9.3; 95% confidence interval, 2.6- 33.9) or a subtle neurologic finding (odds ratio, 8.7; 95% confidence interval, 2.3-33.1). One patient with a new stroke had a prior stroke, 3 were age N65 years, and none had coronary artery disease or dysrhythmia. Among patients with acute stroke, 5 also had Head CT. and none detected the stroke. Conclusions: This study identified 2 variables associated with acute stroke that should be considered in the evaluation of ED patients with vertigo. Head CT was inadequate for diagnosing acute stroke in this patient population.

(C) 2012

Introduction

^? Presented at: Society for Academic Emergency Medicine National Conference, 2009.

* Corresponding author. Tel.: +1 617 754 2298.

E-mail address: [email protected] (M. Chase).

Dizziness and vertigo are common complaints, account- ing for approximately 7.5 million visits to ambulatory care settings annually [1]. Dizziness has historically been classified into 4 categories: vertigo, lightheadedness, pre- syncope, and disequilibrium. Of these 4, vertigo is the most

0735-6757/$ – see front matter (C) 2012 doi:10.1016/j.ajem.2011.02.002

common and is believed to represent roughly 54% of dizziness complaints [2]. Although the large majority of patients with dizziness have benign processes that are not life or brain threatening, a small percentage is due to serious central nervous system pathology including space occupying lesions, infarction, and other processes. Clinically, there is significant overlap in the presentation of patients with peripheral and central causes; symptoms of dizziness, vertigo, nausea, vomiting, imbalance, and headache are all common and nonspecific; and stroke is equally likely to be associated with vertigo as nonvertiginous dizziness [1,3].

For the emergency physician facing a broad range of potential etiologies, there is little research to guide risk stratification for our undifferentiated patient population, and recent data suggest that even neurologists may have difficulty in identifying cerebrovascular disease in the vertebrobasilar system [4]. Distinguishing patients with central processes from those with benign peripheral causes is crucial because patients with a misdiagnosed central nervous system etiology of dizziness have a propensity for bad outcomes. In one case series, half of the patients with missed cerebellar infarctions were less than 50 years, the mortality rate was 40%, and half of all survivors had disabling deficits [5].

As a consequence of this uncertainty and the cost of a missed stroke, patients with dizziness tend to consume greater Health care resources than nondizzy emergency department (ED) patients, including longer lengths of stay in the ED and higher rates of admission, cardiac monitoring, and diagnostic imaging [6,7]. Although head computed tomography is used widely in the ED setting, it has poor sensitivity for brain ischemia or infarction, especially in the posterior fossa. Magnetic resonance imaging (MRI) is the imaging test of choice, but it has several practical limitations, including its poor real-time availability in the emergent setting [8].

At present, there are wide variations in clinical manage- ment of ED patients with vertigo because very little focused data exist in the emergency medicine literature to improve both our understanding and risk stratification of this patient population. We performed this study as an initial step toward this end. We retrospectively reviewed charts of all patients who received an MRI for the indication of vertigo either emergently or within 2 weeks of an ED visit for vertigo. Our goal was to determine the feasibility of creating a risk profile for ED patients with vertigo at risk for acute stroke at time of initial presentation.

Methods

Setting

The study was performed at an Urban academic medical center with an annual ED census of 53 000 patients and Accreditation Council for Graduate Medical Education

(ACGME)-accredited Emergency Medicine and Neurology residency programs. Magnetic resonance imaging is available 24 hours a day for emergency cases. Informed consent was waived via expedited review of the institutional review board at our institution.

Study design

Patients were identified via an electronic query of the MRI patient database for the term vertigo in either the study order or report from January 2007 to June 2008. Using standard methodology, we retrospectively reviewed charts of patients who had an MRI for the indication of vertigo at our institution [9,10]. Study subjects included all patients who presented to the ED with vertigo, with or without additional historical or physical examination findings, prompting an MRI for evaluation either during the index hospitalization or within 2 weeks of the ED visit as an outpatient, as documented in the medical records. Clinical variables of interest were determined a priori based on both past literature and clinical experience of the investigators. The primary outcome, a cerebrovascular cause for vertigo, was confirmed via neurology documentation of acute stroke as the etiology of the patient’s symptoms.

Data collection and analysis

Research assistants trained in data abstraction reviewed the electronic medical record for discharge summaries and Radiology reports. Data recorded onto a standardized data collection form and included demographic information (age, sex, race, and tobacco and illicit drug use), medical history (hypertension, cancer, diabetes, dysrhythmia, coronary artery disease, benign positional vertigo, previous stroke), history of the present illness including Presenting complaints (lightheadedness, dizziness, vertigo, near-syncope, acute stroke), accompanying complaints (gait instability, head- ache, visual disturbance), physical examination findings (nystagmus, focal neurologic finding), and neuroimaging results. Neurology discharge diagnosis was considered definitive diagnosis, and all positive findings on MRI were subsequently confirmed via chart review by 2 study investigators. neurologic findings were characterized as subtle or questionable if they were documented by the neurology consult team either in their own examination or in the indication for the consult (ie, “asked to consult on patient with vertigo and possible facial droop”). Data were entered into an electronic database (Access 2003, Microsoft Corporation, Redmond, WA) and imported into STATA (StataCorp. 2005. Stata Statistical Software: Release 9. College Station, TX) for data analysis. Clinical variables were analyzed for association with the primary outcome, acute stroke identified on MRI, using Fisher’s exact test because of the small cell numbers. Unadjusted odds ratios (ORs) are presented for factors found to have a significant association (P b .05).

Results

There were 325 patients who had an MRI for the indication of vertigo during the study period, of whom 131 (40%) were seen in the ED; 118 of these patients had an MRI during the index hospitalization, and 13 had an MRI within

2 weeks of the ED visit. Non-ED patients (60%) were referred from primary care, neurology, and oncology clinics primarily. Patients were 57 (+-18) years, and 53% were

women (Table 1). Overall, there were 27 patients (8.3%) with a new MRI finding including 12 strokes, 1 dissection, 9 new masses, and 5 other findings (demyelinating lesion, aneurysm, infection).

All 12 patients with a new stroke were ED patients who had an MRI during the index hospitalization (9.2%). Clinical characteristics of these patients are described in Table 1. There were also 4 masses identified in ED patients; other MRI findings occurred in non-ED patients. Among patients with stroke, there were 9 cerebellar and 3 brainstem infarctions. Statistical analysis identified several variables with an association with stroke: a presenting complaint of gait instability (OR, 9.3; 95% confidence interval, 2.6-33.9) or a subtle neurologic finding (OR, 8.7; 95% confidence interval, 2.3-33.1) including positive Romberg test, gait instability, mild facial droop, and dysmetria. Three of our patients had a normal neurologic

Table 1 Characteristics of ED patients with and without stroke

Physical examination findings

Nystagmus present Focal neurologic finding

33

42

20

8

Characteristic	Percentage of ED patients with	Percentage of ED patients without
	stroke (n = 12)	stroke (n = 119)
Female sex	42%	60%
Age N65 y	25	39
Tobacco use	8	13
Medical history Diabetes	8	16
Coronary artery	0	9
disease Hypertension	47	50
Dysrhythmia	0	4
Stroke	8	8
Benign positional vertigo Presenting complaint	0	8
Dizziness	67	54
Vertigo	58	75
Lightheadedness	8	14
Near-syncope	8	6
Headache	25	23
Vision changes	8	5
Gait instability	67	18
Stroke	0	b1

examination with the exception of nystagmus. Details can be found in Table 2.

Only one ED patient with a new stroke had a history of a prior stroke, and only 3 were age N65 years. Furthermore, none had a history of coronary artery disease or dysrhythmia, all of which are classically associated with cerebrovascular risk. Although not significant associations, 8 patients with stroke had a presenting complaint of dizziness and 6 had hypertension (Table 1). Five of the 12 ED patients with stroke also had head CT performed before the MRI, and none of the CT scans showed the stroke.

Discussion

Although a finding of Multiple sclerosis or a new mass is of obvious importance in the evaluation of ED patients with acute vertigo, we focused our investigation on acute stroke because there are clear data that if these patients are not diagnosed, monitored, and treated properly, they may have worse outcomes [3,5].

In our small patient sample, we did not find that traditional Cardiovascular risk factors were associated with a higher risk for stroke. In fact, the average age of our patients was 57 years, and only 3 patients with stroke were older than 65 years. It may be that physicians do not consider the diagnosis of stroke in younger patients; other investiga- tors have found that this may be an independent risk factor for misdiagnosis [5]. Many young patients with cerebellar infarctions have vertebral dissection or cardioembolism (often via a patent foramen ovale) as the underlying vascular mechanism. Misdiagnosed patients will remain at risk for recurrent events.

The complaint of imbalance or difficulty walking has been shown to be a predictor for stroke in patients with dizziness [1,11]. Our investigation supports these prior findings because 8 of our 12 patients with stroke complained of gait instability, and this complaint was one of only 2 variables found to have an association with stroke. Although gait disturbance is not unique to patients with a central cause for their vertigo, abnormal gait is found on physical examination in approximately half of patients with cerebellar infarction [3]. We found an abnormal gait examination in 5 of our 12 ED patients with stroke, although many of our patients did not have a detailed gait examination documen- ted. This is frequently a result of underlying ambulatory dysfunction or an overly symptomatic patient but may represent an underused diagnostic tool.

Recent data suggest that bedside oculomotor examination may actually be more sensitive than MRI [12]. Of our ED patients with stroke, 7 had documented nystagmus. Overall, nystagmus is reported in roughly half of cases of cerebellar stroke, and the finding of direction-changing (fast compo- nent changes with the direction of the patient’s gaze) or vertical nystagmus is thought to suggest a central cause [3].

Age	Sex	Presenting complaints	Nystagmus	Focal neurologic finding	CT head	CT finding of stroke
60	Female	Vertigo, HA, dizzy, gait	Absent	Dysmetria	NO
25	Female	Vertigo, gait, HA	Horizontal	None	NO
56	Male	Dizzy, HA, gait	Absent	Difficulty with tandem gait	YES	NO
89	Male	Dizzy, near-syncope	Absent	Unsteady gait, possible	NO
				positive Romberg test
60	Female	Dizzy	Horizontal and	Mild facial droop, unsteady gait	YES	NO
			torsional
74	Male	Vertigo, gait	Horizontal	None	NO
57	Female	Vertigo, dizzy	Absent	Mild leg weakness	NO
83	Male	Dizzy, gait, visual dist.	Torsional	Positive Romberg test	NO
53	Male	Vertigo	Absent	Positive Romberg test,	NO
				difficulty with tandem gait
53	Male	Dizzy, gait	Vertical	Dysmetria, questionable	YES	Incompletely
				facial droop		characterized
64	Female	Vertigo, dizzy, gait	Torsional	Difficulty with tandem gait	YES	NO
64	Male	Vertigo, dizzy, gait	Torsional	None	YES	NO
HA indicates headache; gait, gait instability; dist., disturbance.

In this study, we did not find a consistent pattern of nystagmus (horizontal, vertical, or torsional) with patients with stroke. However, these examination findings are not a typical part of most physicians’ clinical evaluation, and the presence or absence and character of nystagmus were not documented on all patients with vertigo. Along with better documentation of the gait, this highlights another area for potential clinical improvement to allow emergency physi- cians to better risk-stratify patients with vertigo at risk for posterior circulation stroke.

Table 2 Clinical features and physical examination findings of ED patients with vertigo and acute stroke

Our study identified a subtle neurologic finding on physical examination as a risk factor for acute stroke. Although this may seem intuitive, these are subtle findings that can often be dismissed as normal or baseline level of function for the patient, particularly in the setting of symptomatic vertigo. This uncertainty often prompts emer- gent or urgent neurology consultation and continued testing. A careful history and physical examination targeted at cerebellar and brainstem function may reduce the usage of expensive and time-consuming resources, particularly for those who may be at Very low risk.

Our study has several important limitations. First is the retrospective study design, which created a selection bias toward those patients with vertigo who were selected for further imaging. This factor would likely lead us to overestimate the incidence of cerebrovascular pathology in this population of patients. It is also possible that some patients with milder vertigo also caused by acute stroke went undiagnosed during the study period. Retrospective analysis also limited our ability to collect specific data points, forcing a reliance on documented examination findings that may have lacked detail.

Furthermore, we used any MRI done within 2 weeks of the index ED visit. Although unlikely, it is possible that some

patients fell outside this window. Last, the small sample size and low distribution of the variables of interest among patients with the stroke limit our ability to draw definitive conclusions about associations between these factors and our outcome.

Conclusion

Acute posterior circulation infarction can present with common, nonspecific clinical findings and a normal head CT. Despite the limitations of our retrospective study, we were able to identify 2 clinical features in our patient population associated with acute stroke. These findings underscore that a thorough history and physical examination focused on cerebellar and posterior circulation functions performed in real time at the bedside are keys to the diagnostic workup for patients with vertigo. We hypothesize that increased education about and focus on these features will allow for appropriate identification of high-risk patients who warrant emergent MRI to exclude acute stroke. Clearly, more focused prospective research with a larger patient population is needed to definitively identify risk factors for posterior circulation stroke, such that health care resources can be conserved for those at highest risk.

References

1. Kerber KA, Brown DL, Lisabeth LD, et al. Stroke among patients with dizziness, vertigo, and imbalance in the emergency department: a population-based study. Stroke 2006;37(10):2484-7.
2. Kroenke K, Hoffman RM, Einstadter D. How common are various causes of dizziness? A critical review. South Med J 2000;93(2):160-7 quiz 168.
3. Edlow JA, Newman-Toker DE, Savitz SI. Diagnosis and initial manage- ment of cerebellar infarction. Lancet Neurol 2008;7(10):951-64.
4. Flossmann E, Redgrave JN, Briley D, et al. Reliability of clinical diagnosis of the symptomatic vascular territory in patients with recent transient ischemic attack or minor stroke. Stroke 2008;39(9):2457-60.
5. Savitz SI, Caplan LR, Edlow JA. Pitfalls in the diagnosis of cerebellar infarction. Acad Emerg Med 2007;14(1):63-8.
6. Newman-Toker DE, Hsieh YH, Camargo Jr CA, et al. Spectrum of dizziness visits to US emergency departments: cross-sectional analysis from a nationally Representative sample. Mayo Clin Proc 2008;83(7): 765-75.
7. Stewart MG, Chen AY, Wyatt JR, et al. Cost-effectiveness of the diagnostic evaluation of vertigo. Laryngoscope 1999;109(4):600-5.
8. Chalela JA, Kidwell CS, Nentwich LM, et al. Magnetic resonance imaging and computed tomography in emergency assessment of

patients with suspected acute stroke: a prospective comparison. Lancet 2007;369(9558):293-8.

1. Gilbert EH, Lowenstein SR, Koziol-McLain J, et al. Chart reviews in emergency medicine research: where are the methods? Ann Emerg Med 1996;27(3):305-8.
2. Worster A, Bledsoe RD, Cleve P, et al. Reassessing the methods of medical record review studies in emergency medicine research. Ann Emerg Med 2005;45(4):448-51.
3. Lee H, Sohn SI, Cho YW, et al. Cerebellar infarction presenting isolated vertigo: frequency and vascular topographical patterns. Neurology 2006;67(7):1178-83.
4. Kattah JC, Talkad AV, Wang DZ, et al. HINTS to diagnose stroke in the acute vestibular syndrome: three-step bedside oculomotor examination more sensitive than early MRI diffusion-weighted imaging. Stroke 2009;40(11):3504-10.