Eye Signs Flag Stroke in Vestibular Syndrome

BALTIMORE — A test battery of three oculomotor signs for patients with acute vestibular syndrome can detect stroke with greater sensitivity than does MRI with diffusion-weighted imaging in the first 24-48 hours after symptom onset, according to a prospective, cross-sectional study.

If the results are confirmed in a larger, multicenter study, they will help determine which patients with acute vestibular syndrome (AVS) should undergo MRI scanning, Dr. David E. Newman-Toker said at the annual meeting of the American Neurological Association.

MRI scanning of all patients with AVS is “probably an unrealistic strategy” for diagnosis, said Dr. Newman-Toker of the departments of neurology and ophthalmology at Johns Hopkins University, Baltimore.

Of 2.6 million visits to the emergency department for dizziness each year, about 5% (or 100,000-150,000 people) have a cerebrovascular event, mostly in the lateral brainstem or the inferior cerebellum (Mayo Clin. Proc. 2008;83:765-75). Of that 5%, evidence suggests that approximately 35% may be misdiagnosed, which is a much higher rate than with other types of stroke (Stroke 2006;37:2484-7). Another study reported that around 40% of these misdiagnoses culminate in death or disability.

Strokes in the lateral brainstem or the inferior cerebellum frequently mimic benign vestibular disorders, such as vestibular neuritis or labyrinthitis, which are collectively known as acute peripheral vestibulopathies.

To determine if a battery of bedside oculomotor signs could detect patients with stroke with greater accuracy than did early MRI with diffusion-weighted imaging (DWI), the investigators prospectively enrolled 101 patients during 1999-2008 who presented with AVS to an urban academic acute stroke referral center. AVS is a rapid onset of a new persistent dizziness or vertigo that is sustained for 12-24 hours with some degree of gait unsteadiness, in association with head-motion intolerance and nystagmus, nausea, and vomiting.

The patients were admitted either through the academic medical center's emergency department or by transfer from other hospitals. They only included patients who had at least one risk factor for stroke.

Each patient underwent MRI scanning with DWI, which was repeated if the initial MRI was negative but oculomotor and other neurologic signs indicated stroke. “This distinguishes this study from every other study to date on this subject, where it was assumed that the initial MRI scan in those studies was accurate, which was not necessarily a safe assumption,” Dr. Newman-Toker said.

One investigator performed the test battery for oculomotor signs of stroke—given the mnemonic HINTS: head impulse test, nystagmus, and test of skew.

These are assessed by performing a horizontal head impulse test (a test of vestibulo-ocular reflex function), a test for direction-changing nystagmus, and an alternating cover test to test for skew deviation. The results of each test have been individually associated with stroke in AVS (Stroke 2009;40:3504-10).

Men accounted for 65% of the enrolled patients, who had a mean age of 62 years. Most patients (70%) had two or more stroke risk factors, “so this was a very high risk for stroke population,” he noted.

Three-fourths of the patients were examined within 24 hours of symptom onset and about 70% underwent an MRI scan within 6 hours of their bedside exam. A single examiner, who was blinded only to the MRI findings, performed all of the neurologic exams.

Based on the reference standard of MRI with DWI for diagnosis, 25 patients had peripheral vestibulopathy (vestibular neuritis or labyrinthitis) and 76 had central brain pathology (69 ischemic strokes, 4 hemorrhages, 2 demyelinating diseases, and 1 acute intoxication with an anticonvulsant).

Testing for HINTS to INFARCT (Impulse Normal, Fast-Phase Alternating, or Refixation on Cover Test) detected stroke with 100% sensitivity and 96% specificity. In contrast, an initial MRI with DWI detected stroke with 88% sensitivity and 100% specificity. The use of any obvious neurologic signs for detecting stroke (such as limb ataxia, severe truncal ataxia, hemiparesis, or gaze palsy) provided a sensitivity of 64% with 100% specificity. General neurologic signs had only 19% sensitivity and 100% specificity. A normal horizontal head impulse test was the best single predictor of stroke with 90% sensitivity and 100% specificity, although it misses lateral pontine strokes because the reflex pathway tracks from the inner ear straight to the pons.

Peripheral vestibulopathies occurred in patients with an abnormal head impulse test, direction-fixed nystagmus, and no skew deviation.

The study was funded by grants from the National Institutes of Health and the Agency for Healthcare Research and Quality. Dr. Newman-Toker reported having no relevant disclosures.

BALTIMORE — A test battery of three oculomotor signs for patients with acute vestibular syndrome can detect stroke with greater sensitivity than does MRI with diffusion-weighted imaging in the first 24-48 hours after symptom onset, according to a prospective, cross-sectional study.

If the results are confirmed in a larger, multicenter study, they will help determine which patients with acute vestibular syndrome (AVS) should undergo MRI scanning, Dr. David E. Newman-Toker said at the annual meeting of the American Neurological Association.

MRI scanning of all patients with AVS is “probably an unrealistic strategy” for diagnosis, said Dr. Newman-Toker of the departments of neurology and ophthalmology at Johns Hopkins University, Baltimore.

Of 2.6 million visits to the emergency department for dizziness each year, about 5% (or 100,000-150,000 people) have a cerebrovascular event, mostly in the lateral brainstem or the inferior cerebellum (Mayo Clin. Proc. 2008;83:765-75). Of that 5%, evidence suggests that approximately 35% may be misdiagnosed, which is a much higher rate than with other types of stroke (Stroke 2006;37:2484-7). Another study reported that around 40% of these misdiagnoses culminate in death or disability.

Strokes in the lateral brainstem or the inferior cerebellum frequently mimic benign vestibular disorders, such as vestibular neuritis or labyrinthitis, which are collectively known as acute peripheral vestibulopathies.

To determine if a battery of bedside oculomotor signs could detect patients with stroke with greater accuracy than did early MRI with diffusion-weighted imaging (DWI), the investigators prospectively enrolled 101 patients during 1999-2008 who presented with AVS to an urban academic acute stroke referral center. AVS is a rapid onset of a new persistent dizziness or vertigo that is sustained for 12-24 hours with some degree of gait unsteadiness, in association with head-motion intolerance and nystagmus, nausea, and vomiting.

The patients were admitted either through the academic medical center's emergency department or by transfer from other hospitals. They only included patients who had at least one risk factor for stroke.

Each patient underwent MRI scanning with DWI, which was repeated if the initial MRI was negative but oculomotor and other neurologic signs indicated stroke. “This distinguishes this study from every other study to date on this subject, where it was assumed that the initial MRI scan in those studies was accurate, which was not necessarily a safe assumption,” Dr. Newman-Toker said.

One investigator performed the test battery for oculomotor signs of stroke—given the mnemonic HINTS: head impulse test, nystagmus, and test of skew.

These are assessed by performing a horizontal head impulse test (a test of vestibulo-ocular reflex function), a test for direction-changing nystagmus, and an alternating cover test to test for skew deviation. The results of each test have been individually associated with stroke in AVS (Stroke 2009;40:3504-10).

Men accounted for 65% of the enrolled patients, who had a mean age of 62 years. Most patients (70%) had two or more stroke risk factors, “so this was a very high risk for stroke population,” he noted.

Three-fourths of the patients were examined within 24 hours of symptom onset and about 70% underwent an MRI scan within 6 hours of their bedside exam. A single examiner, who was blinded only to the MRI findings, performed all of the neurologic exams.

Based on the reference standard of MRI with DWI for diagnosis, 25 patients had peripheral vestibulopathy (vestibular neuritis or labyrinthitis) and 76 had central brain pathology (69 ischemic strokes, 4 hemorrhages, 2 demyelinating diseases, and 1 acute intoxication with an anticonvulsant).

Testing for HINTS to INFARCT (Impulse Normal, Fast-Phase Alternating, or Refixation on Cover Test) detected stroke with 100% sensitivity and 96% specificity. In contrast, an initial MRI with DWI detected stroke with 88% sensitivity and 100% specificity. The use of any obvious neurologic signs for detecting stroke (such as limb ataxia, severe truncal ataxia, hemiparesis, or gaze palsy) provided a sensitivity of 64% with 100% specificity. General neurologic signs had only 19% sensitivity and 100% specificity. A normal horizontal head impulse test was the best single predictor of stroke with 90% sensitivity and 100% specificity, although it misses lateral pontine strokes because the reflex pathway tracks from the inner ear straight to the pons.

Peripheral vestibulopathies occurred in patients with an abnormal head impulse test, direction-fixed nystagmus, and no skew deviation.

The study was funded by grants from the National Institutes of Health and the Agency for Healthcare Research and Quality. Dr. Newman-Toker reported having no relevant disclosures.

BALTIMORE — A test battery of three oculomotor signs for patients with acute vestibular syndrome can detect stroke with greater sensitivity than does MRI with diffusion-weighted imaging in the first 24-48 hours after symptom onset, according to a prospective, cross-sectional study.

If the results are confirmed in a larger, multicenter study, they will help determine which patients with acute vestibular syndrome (AVS) should undergo MRI scanning, Dr. David E. Newman-Toker said at the annual meeting of the American Neurological Association.

MRI scanning of all patients with AVS is “probably an unrealistic strategy” for diagnosis, said Dr. Newman-Toker of the departments of neurology and ophthalmology at Johns Hopkins University, Baltimore.

Of 2.6 million visits to the emergency department for dizziness each year, about 5% (or 100,000-150,000 people) have a cerebrovascular event, mostly in the lateral brainstem or the inferior cerebellum (Mayo Clin. Proc. 2008;83:765-75). Of that 5%, evidence suggests that approximately 35% may be misdiagnosed, which is a much higher rate than with other types of stroke (Stroke 2006;37:2484-7). Another study reported that around 40% of these misdiagnoses culminate in death or disability.

Strokes in the lateral brainstem or the inferior cerebellum frequently mimic benign vestibular disorders, such as vestibular neuritis or labyrinthitis, which are collectively known as acute peripheral vestibulopathies.

To determine if a battery of bedside oculomotor signs could detect patients with stroke with greater accuracy than did early MRI with diffusion-weighted imaging (DWI), the investigators prospectively enrolled 101 patients during 1999-2008 who presented with AVS to an urban academic acute stroke referral center. AVS is a rapid onset of a new persistent dizziness or vertigo that is sustained for 12-24 hours with some degree of gait unsteadiness, in association with head-motion intolerance and nystagmus, nausea, and vomiting.

The patients were admitted either through the academic medical center's emergency department or by transfer from other hospitals. They only included patients who had at least one risk factor for stroke.

Each patient underwent MRI scanning with DWI, which was repeated if the initial MRI was negative but oculomotor and other neurologic signs indicated stroke. “This distinguishes this study from every other study to date on this subject, where it was assumed that the initial MRI scan in those studies was accurate, which was not necessarily a safe assumption,” Dr. Newman-Toker said.

One investigator performed the test battery for oculomotor signs of stroke—given the mnemonic HINTS: head impulse test, nystagmus, and test of skew.

These are assessed by performing a horizontal head impulse test (a test of vestibulo-ocular reflex function), a test for direction-changing nystagmus, and an alternating cover test to test for skew deviation. The results of each test have been individually associated with stroke in AVS (Stroke 2009;40:3504-10).

Men accounted for 65% of the enrolled patients, who had a mean age of 62 years. Most patients (70%) had two or more stroke risk factors, “so this was a very high risk for stroke population,” he noted.

Three-fourths of the patients were examined within 24 hours of symptom onset and about 70% underwent an MRI scan within 6 hours of their bedside exam. A single examiner, who was blinded only to the MRI findings, performed all of the neurologic exams.

Based on the reference standard of MRI with DWI for diagnosis, 25 patients had peripheral vestibulopathy (vestibular neuritis or labyrinthitis) and 76 had central brain pathology (69 ischemic strokes, 4 hemorrhages, 2 demyelinating diseases, and 1 acute intoxication with an anticonvulsant).

Testing for HINTS to INFARCT (Impulse Normal, Fast-Phase Alternating, or Refixation on Cover Test) detected stroke with 100% sensitivity and 96% specificity. In contrast, an initial MRI with DWI detected stroke with 88% sensitivity and 100% specificity. The use of any obvious neurologic signs for detecting stroke (such as limb ataxia, severe truncal ataxia, hemiparesis, or gaze palsy) provided a sensitivity of 64% with 100% specificity. General neurologic signs had only 19% sensitivity and 100% specificity. A normal horizontal head impulse test was the best single predictor of stroke with 90% sensitivity and 100% specificity, although it misses lateral pontine strokes because the reflex pathway tracks from the inner ear straight to the pons.

Peripheral vestibulopathies occurred in patients with an abnormal head impulse test, direction-fixed nystagmus, and no skew deviation.

The study was funded by grants from the National Institutes of Health and the Agency for Healthcare Research and Quality. Dr. Newman-Toker reported having no relevant disclosures.