Original Article
Stroke Chameleons

https://doi.org/10.1016/j.jstrokecerebrovasdis.2013.07.015Get rights and content

Background

Many conditions called “stroke mimics” may resemble acute stroke. The converse of the “stroke mimic” is a presentation suggestive of another condition, which actually represents stroke. These would be “stroke chameleons.” The recognition of a chameleon as stroke has implications for therapy and quality of care.

Methods

We performed a retrospective chart review, including all cases for 1 year in which patients had a stroke missed on hospital presentation. Initial erroneous diagnoses were compared for all patients correctly admitted with those diagnoses to determine positive predictive value (PPV) for each chameleon.

Results

Ninety-four cases were identified as chameleons where brain imaging revealed acute stroke. The common chameleons were initially diagnosed as altered mental status (AMS) (29, 31%), syncope (15, 16%), hypertensive emergency (12, 13%), systemic infection (10, 11%), and suspected acute coronary syndrome (ACS) (9, 10%). The total number of patients who were diagnosed with these conditions over the same year were AMS (393), syncope (326), hypertensive emergency (144), systemic infection (753), and suspected ACS (817) (total N = 2528). For each chameleon diagnosis, the PPV of each presentation for acute stroke was AMS (7%), syncope (4%), hypertensive emergency (8%), systemic infection (1%), and suspected ACS (1%).

Conclusions

Stroke chameleons may result in patients not receiving appropriate care. The largest proportions of chameleons were AMS, syncope, hypertensive emergency, systemic infection, and suspected ACS. Patients diagnosed with hypertensive emergency or AMS had an 8% and 7% chance of having an acute stroke. Physicians should consider stroke in patients with these diagnoses with a lower threshold to obtain neuroimaging with subsequent appropriate management.

Introduction

Stroke is most clearly characterized by the sudden onset of an acute neurologic deficit. We coined the term "stroke mimic" to describe a clinical syndrome suggestive of stroke, yet not actually caused by an ischemic event.1 In a prior study, we showed that of approximately 400 patients diagnosed as stroke, the prevalence of stroke mimics was found to be 19%. These conditions were found to primarily include seizures (17%), systemic infections (17%), brain tumor (15%), and toxic metabolic disturbances (13%). The rate of stroke mimics in other studies has been as low as 4% with the addition of computed tomography and laboratory analysis.2

As opposed to stroke mimics, a related, but not well-explored concept is what may be called a “stroke chameleon.”3 This concept encompasses syndromes that do not appear to represent a stroke on initial presentation but are later found to represent an acute stroke. Stroke chameleons may present as malaise, loss of consciousness, encephalopathy, and acute psychosis and a myriad of other conditions. Although symptoms such as weakness, sensory loss, and hemianopia are easy to recognize, the more subtle presentations of stroke may be easily missed. Abulia, for instance, is well documented in anterior cerebral arterial vascular distribution infarcts. Even more subtly, simple confusion is well documented in infarcts of multiple areas including thalamus, nondominant parietal lobe, and caudate nucleus. Cases of peduncular hallucinosis, for example, are well described as a consequence of stroke and symptoms may easily be attributed to an acute psychiatric event. Thalamic infarcts may have presentations indistinguishable from schizophrenia (with the exception of the acuity of onset).4

The rapid recognition of a chameleon as an acute stroke has significant clinical implications, in terms of acute therapy, secondary prevention, and overall quality improvement of stroke care. This vigilance may assist with compliance with performance improvement instruments such as “Get with the Guidelines”5 and result in improved patient care. Clarification of the most common stroke chameleons would allow clinicians to devote appropriate resources to investigate and treat patients for stroke, who might otherwise be missed. In patients determined to be at a higher risk for having had an acute stroke, it may be advantageous to have a lower threshold for brain imaging followed by appropriate workup.

Section snippets

Methods

A retrospective chart review was performed on all cases from January 2011 to January 2012, in whom an acute stroke was missed on initial hospital presentation. Data were pulled from the administrative database of a state Department of Health–certified primary care stroke center. All such cases then had a full chart review before inclusion or subsequent exclusion. Included were all patients presenting with an acute syndrome leading to hospital admission and given an initial admitting diagnosis

Results

Among the 94 patients found to present with stroke chameleons, the 5 most common were altered mental status (AMS), syncope, hypertensive emergency, systemic infection, and suspected acute coronary syndrome (ACS). Figure 1 shows the breakdown of these stroke chameleons. These stroke chameleons and their PPVs are depicted in Table 1. A list of the remaining chameleons is summarized in Table 2. In the same 1-year period, a total of 2528 patients were admitted with the same diagnoses as the stroke

Discussion

Stroke mimics and stroke chameleons present a diagnostic challenge to the clinician evaluating patients in the acute setting. The most common stroke chameleon in our study was AMS. Presentations labeled as AMS included acute psychosis, perceived delirium, memory deficits, and generalized encephalopathy. Benbaddis et al6 found that 8 of 127 patients (6%) seen on consultation for isolated mental status change were found to have an acute ischemic stroke, similar to that found in our study. In

References (15)

  • J.S. Huff

    Stroke mimics and chameleons

    Emerg Med Clin N Am

    (2002)
  • R.B. Libman et al.

    Conditions that mimic stroke in the emergency department. Implications for acute stroke trials

    Arch Neurol

    (1995)
  • R.U. Kothari et al.

    Emergency physicians: accuracy in diagnosis of stroke

    Stroke

    (1995)
  • E. Carrera et al.

    The thalamus and behavior: effects of anatomically distinct strokes

    Neurology

    (2006)
  • K. LaBresh et al.

    Hospital treatment of patients with ischemic stroke or transient ischemic attack with Get With The Guidelines

    Arch Intern Med

    (2008)
  • S. Benbaddis et al.

    Mental status changes and stroke

    J Stroke Cerebrovasc Dis

    (1994)
  • J. Parvizi et al.

    Neuroanatomical correlates of brainstem coma

    Brain

    (2003)
There are more references available in the full text version of this article.

Cited by (49)

  • Neurology

    2023, Ketogenic: The Science of Therapeutic Carbohydrate Restriction in Human Health
  • Neuroimaging in acute ischemic stroke: Trends, disparities, and clinical impact

    2022, European Journal of Radiology
    Citation Excerpt :

    The association between early clinical outcomes (good discharge mRS 0–2 and in-hospital mortality/hospice discharge) and patient characteristics, imaging, treatment, and year were studied in separate regression models. In these analyses, unknown admission NIHSS patients were grouped with the < 6 category because patients discharged from our CSC without admission NIHSS typically present as stroke chameleons [21] with low NIHSS. Missing data for LKWA was grouped with the > 24-hour category, as these patients are managed similarly in terms of imaging and treatment.

  • General Stroke Management and Stroke Units

    2021, Stroke: Pathophysiology, Diagnosis, and Management
View all citing articles on Scopus

Contributions: C.M.D. performed all chart reviews, participated in data review and interpretation, and drafted the initial manuscript and subsequent revisions. R.L. is responsible for study mentorship and conceptualization and also data review and revision of the manuscript for intellectual content. S.I.D. performed all statistics, tables, and figures and was instrumental in overall study design. J.M.K., I.R., and T.K. are responsible for editorial review and making significant contribution in revising the manuscript for intellectual content.

Disclosures: This study was unsponsored, and all authors report no disclosures.

View full text