From the Department of Neurology (C.A.C.W, P.A.W, C.S.K, M.K.-H.), Boston
University School of Medicine, and the Department of Epidemiology and
Biostatistics (A.S.B.), Boston University School of Public Health (Mass), and
the Framingham Study, National Heart, Lung, and Blood Institute, Framingham,
Mass.
Correspondence to Philip A. Wolf, MD, Department of Neurology, Boston University School of Medicine, B608, 715 Albany St, Boston, MA 02118. E-mail pawolf{at}bu.edu
Methods—During 1971–1989, at biennial examinations, 2110
subjects of the Framingham cohort were systematically queried about the
occurrence of sudden visual symptoms.
Results—Visual migrainous symptoms were reported by 1.23%
(26/2110) of subjects (1.33% of women and 1.08% of men). In 65% of
subjects the episodes were stereotyped, and they began after age 50
years in 77%. Mean±SD age at onset of the episodes was 56.2±18.7
years. In 58% of subjects the episodes were never accompanied by
headaches, and 42% had no headache history. The number of episodes
ranged from 1 to 500 and was 10 or more in 69% of subjects. The
episodes lasted 15 to 60 minutes in 50% of subjects. Sixty-five
percent of the subjects were examined by a study neurologist, and only
19% of them met the criteria of the International Headache Society.
Twelve percent of subjects sustained a stroke after the onset of
migrainous visual symptoms: a subarachnoid hemorrhage 1
year later, an atherothrombotic brain stem infarct 3 years later, and a
cardioembolic stroke 27 years later. In contrast, of 87 subjects with
TIAs in the same cohort, 33% developed a stroke
(P=0.030), two thirds within 6 months of TIA onset.
Conclusions—Late-life-onset transient visual phenomena similar to
the visual aura of migraine are not rare and often occur in the absence
of headache. These symptoms appear not to be associated with an
increased risk of stroke, and invasive diagnostic
procedures or therapeutic measures are generally not indicated.
Our knowledge of migraine accompaniments is largely derived from
clinical series.1 2 3 9 11 Visual symptoms are the
most frequent manifestation.1 3 6 Typical
migrainous visual symptoms include both positive (scintillations,
fortification spectra, photopsia) and negative (scotoma, hemianopsia)
visual features6 13 14 and often present as a
transient scintillating hemianopic visual
disturbance.6 15 Isolated migraine
accompaniments may occur with a prior, remote history of migraine
headaches with or without aura. On the basis of his detailed clinical
observations, Fisher1 described a number of
features of migraine accompaniments, including scintillations and other
visual phenomena, buildup of scintillations and march of paresthesias,
and progression of one neurological symptom to another. Other
characteristics that he described were episodes lasting 15 to 25
minutes, two or more identical spells, a mid-life "flurry" of
spells, and a benign course.
Diagnosis is not difficult when the spells are typically migrainous in
character and two or more stereotypic spells have occurred. However,
when the spells are atypical or when they occur for the first time in
adult life in a person without a migraine history, other neurological
diagnoses such as TIAs or seizures are a consideration. How frequently
isolated migraine accompaniments occur is presently unknown. Since
data on migraine accompaniments derived from population-based studies
are sparse, this study was undertaken to further clarify the
prevalence, characteristics, and outcome of visual migrainous symptoms
in the Framingham Study population.
For the present study the nature of all first-ever episodes of
sudden visual defect (SVD) reported at biennial examinations during
1971–1989 (examinations 12 through 20) was determined by a neurologist
based on the available notes of the biennial examination, neurology
clinic notes, records from hospitalizations and office visits, and
previous reviews of study investigators (described above). Whenever the
clinical features of the SVD corresponded to the characteristics of the
visual aura of migraine headaches,15 a number of
variables were systematically recorded, including details on
the character and number of episodes, their duration and mode of onset,
accompanying headache, associated symptoms (neurological and
nonneurological), and family history. In addition, it was determined
whether the episodes fulfilled IHS criteria for migraine with aura or
migraine aura without headache.12
Prevalence of visual migraine accompaniments was determined in all
subjects who attended at least six of the nine biennial examinations
during 1971–1989. Thus, each subject was queried on at least six
different occasions about the occurrence of sudden visual symptoms over
an 18-year period. Stroke incidence rates between groups with and
without SVD were compared with the
Ocular diseases resulted in SVD in 17% (32/186) of subjects, who
typically presented with loss of vision, often persistent,
affecting one eye. According to the report of the subject, subsequent
ophthalmologic evaluation had identified an ocular cause such as
retinal hemorrhage, retinal detachment, central retinal artery
occlusion, and acute glaucoma. The largest single category of SVDs was
due to stroke, TIA, or transient monocular blindness, which occurred in
34% (64/186) of subjects. SVD resulting from "other causes"
included diagnoses such as presyncope, seizures, and temporal arteritis
and occurred in 12% of subjects. Visual symptoms in this group varied
widely and usually were associated with a number of other (nonvisual)
symptoms or signs suggestive for the diagnosis. No definite cause for
the reported SVD was found in 22% of subjects. These subjects
frequently reported nonspecific visual symptoms such as blurring of
vision affecting one or both eyes for which no cause was found.
Visual symptoms that corresponded to the visual aura of migraine were
reported by 26 of 186 subjects (14%). The pattern of visual
manifestations varied widely among subjects, and some of the more
detailed descriptions are shown in Table 2
Details on the clinical features of the episodes, personal
characteristics of the subjects, and their medical and family histories
were recorded, whenever available (Table 3
Sixty-five percent of subjects were interviewed and examined by a study
neurologist, and the clinical records of 81% were evaluated by a
panel of investigators including a neurologist. The episodes of
migrainous visual symptoms were always transient, and persistent visual
deficits were not reported. Formal ophthalmologic examination was
documented as normal in three subjects. All subjects that were examined
by one of the study neurologists had full visual fields to
confrontation and sharp discs on funduscopic examination. Neuroimaging
studies were not routinely obtained in subjects in this study, largely
because of the presumed diagnosis of migraine. Furthermore, CT imaging
was not available in Framingham until 1978. Only in 19% of subjects
did the migrainous visual episodes meet the criteria for migraine with
aura or migraine aura without headache of the IHS, usually because one
of the criteria ("at least one aura symptom develops gradually over
more than 4 minutes") could not be reliably ascertained.
Three of 26 subjects (11.5%) sustained a stroke after the onset of
episodes of migrainous visual phenomena: a subarachnoid
hemorrhage 1 year after symptom onset (with negative carotid
angiogram), an atherothrombotic brain stem infarct 3 years after
symptom onset, and a cardioembolic stroke in the setting of atrial
fibrillation 27 years after symptom onset. This stroke incidence rate
of 11.5% was significantly lower than the stroke incidence rate of
33.3% in subjects with TIAs in the same cohort (P=0.030)
and did not differ from the stroke incidence rate of 13.6% of those
without migrainous phenomena or TIAs (Table 4
The prevalence of migraine equivalents in our study may well be
an underestimate because our case ascertainment aimed at identifying
subjects with visual symptoms. Although data from clinical series
indicate that the vast majority of subjects with migraine
accompaniments experience visual phenomena as part of the symptom
complex, some subjects only experience nonvisual neurological
disturbances during the attack.1 3 These
subjects were not identified in our study. In addition, in a few
subjects who reported an SVD, migraine accompaniments were suspected,
but a reliable diagnosis could not be established because of limited
documentation of the details of the episode. Furthermore, a number of
subjects with visual migraine accompaniments may have recognized these
episodes as being similar to the visual aura that they experienced
earlier in life in association with migraine headaches and may not have
reported their symptoms as an SVD.
There are several clinical series of patients reporting symptoms of
migraine aura in the absence of
headache.1 2 3 9 11
Fisher1 3 described a total of 205 cases of
migrainous accompaniments in subjects older than 40 years in two
reports. He excluded subjects with exclusively scintillating visual
phenomena, stating that these episodes were well known to
physicians.1 Headache occurred in association
with the spells in 40% to 50% of cases, and a history of recurrent
headache was present in 50% to 65%.1 3
O'Conner and Tredici9 described 61 cases, all
men, seen during a 15-year period at the US Air Force School of
Aerospace Medicine. These cases were derived from a selected group of
highly trained young men whose profession requires outstanding visual
abilities. Age of onset of spells was 12 to 44 years. Family history
was present in 15 (24.6%) of the cases, and a history of migraine
was present only in 2 (3.3%). Eighteen subjects (29.5%)
experienced neurological deficits other than visual phenomena during
the episodes. Permanent neurological deficit occurred in 1 patient.
Cohen et al2 reported 31 cases of transient
visual phenomena attributed to migraine. Headache was present in 20
patients (64.5%). Sixty-one percent of cases had a positive family
history, and 57% had a personal history of migraine. During a mean
follow-up of 2.2 years, 1 patient had died of cardiac disease, none
suffered a stroke, 1 developed amaurosis fugax, and 1 developed
transient global amnesia. Whitty11 described 16
cases, 7 women and 9 men. Headaches occurred with some of the spells in
9 cases (56%). None of them developed a persistent neurological
deficit. Wiley21 reported 10 patients, 7 women
and 3 men, with scintillating scotomas without headache. Eight of the
10 had no family history of migraine, and none developed persistent
neurological sequelae after an average follow-up of 1.5 years.
The findings of our study are comparable to those described in these
clinical series. Mean age of onset of spells was 56.2 years in our
study. Headache was present with all spells in 30% and with some
spells in 12% of cases. Fifty-four percent of cases had a history of
recurrent headaches. A family history of migraine headaches was
present in 72% of subjects with a documented family history.
One of the characteristic features that distinguish migraine
accompaniments from TIAs is a benign course in the
former.1 In one study,22 vascular
events occurred in 1 of 50 cases (2%) of "migraine aura without
headache" compared with 5 of 50 (10%) in age-matched controls with
TIA. However, there are occasional reports of
ischemic4 23 and
hemorrhagic24 cerebral infarction associated with
migraine accompaniments. In addition, migraine has been identified as
an independent risk factor for ischemic stroke in men older
than 40 years25 and in women younger than 45
years.26 27 28 29 30 However, the absolute risk of stroke
associated with migraine is small.31 In this
study we did not find an increased stroke risk in association with
migrainous visual symptoms, confirming the experience derived from
previous clinical series.
Study Strengths and Limitations
Conclusions and Clinical Implications
Received April 2, 1998;
revision received May 13, 1998;
accepted May 13, 1998.
© 1998 American Heart Association, Inc.
Original Contributions
Migrainous Visual Accompaniments Are Not Rare in Late Life
The Framingham Study
Abstract
Top
Abstract
Introduction
Subjects and Methods
Results
Discussion
References
Background and
Purpose—Questionnaires to elicit symptoms of transient
ischemic attacks (TIAs) may detect late-life transient visual
symptoms similar to the visual aura of migraine, often without
headache. We determined the frequency, characteristics, and stroke
outcome of these symptoms in the Framingham Study.
Key Words: epidemiology • migraine • vision disorders
Introduction
Top
Abstract
Introduction
Subjects and Methods
Results
Discussion
References
Questionnaires used
in epidemiological studies to identify symptoms of transient
ischemic attacks (TIAs) may detect transient visual phenomena
similar to the aura of migraine, often in the absence of headache.
Fisher1 emphasized the benign nature of these
episodes and wrote in 1980: "It is an old observation that
scintillating scotomas may appear for the first time in mid-life in the
absence of a history of migraine or after a migraine-free period of
many years." These episodes have been recognized as migrainous by
several others and have been referred to as "migraine
accompaniments," "acephalgic migraine," and "migraine
equivalents."2 3 4 5 6 7 8 9 10 11 The International Headache
Society (IHS) classified these episodes as "migraine aura without
headache" and published diagnostic criteria in
1988.12
Subjects and Methods
Top
Abstract
Introduction
Subjects and Methods
Results
Discussion
References
The Framingham cohort of 5070 men and women between 30 and 62
years of age and free of cardiovascular disease,
including TIA and stroke at entry, has been followed by means of
routine examinations every 2 years from 1950 until the present.
Conduction of the study has been approved by the Institutional Review
Board of Boston University, and all participants have given informed
consent for their participation in the study. Details of the study
design, implementation, and diagnosis criteria have been published
previously.16 At each biennial examination,
subjects have been systematically queried about the occurrence of
symptoms of TIA and stroke. Since 1968 all subjects were specifically
asked about the following symptoms: unconsciousness, sudden difficulty
with speech, sudden muscular weakness, numbness or tingling, double
vision, loss of vision in one eye, and other sudden visual symptoms.
The records of subjects responding affirmatively to any of these
questions and with suspected symptoms of stroke or TIA were reviewed by
a panel of investigators including a neurologist, and a diagnosis was
established on the basis of detailed clinical, laboratory, and
radiological data. Whenever symptoms were atypical or the available
clinical information was insufficient to make a diagnosis with
certainty, subjects were interviewed and examined by one of the study
neurologists. In addition, stroke surveillance was maintained by daily
monitoring of hospital admissions to the only general hospital in town,
and since 1968, whenever possible, a study neurologist has examined
subjects with TIA or stroke at the time of the hospitalization. 
2 test. All
testing was performed at a 5% significance level.
Results
Top
Abstract
Introduction
Subjects and Methods
Results
Discussion
References
Among the 2110 original cohort members who attended at least six
biennial examinations during 1971–1989, 186 reported a first-ever SVD.
Thus, the incidence of SVD in this population sample was 8.8%. With
the use of all available information, a judgment was made about the
cause of the SVDs in these 186 subjects (Table 1
).
View this table:
[in a new window]
Table 1. Cause of First-Ever Sudden Visual Defect Reported by
186 Subjects Who Attended at Least Six Framingham Study Biennial
Examinations (1971–1989) . The prevalence of migrainous visual
symptoms in this general population was 1.23% overall (1.33% in women
and 1.08% in men). The prevalence of episodes of migrainous visual
symptoms that were never accompanied with headache was 0.71% overall
(0.70% in women and 0.72% in men).
View this table:
[in a new window]
Table 2. Some of the Patterns of Migrainous Visual
Disturbances Described by Subjects in This
Study ). In the majority of subjects
the episodes were stereotyped (65%). The age of onset of the episodes
ranged from 9 to 71 years and was greater than 50 years in 77% of
subjects. Mean±SD age of onset was 56.2±18.7 years overall
(58.1±16.3 years in women and 52.9±23.2 years in men). In 58% of
subjects visual symptoms were never accompanied by headaches, and 12%
experienced headaches with the spells occasionally. Forty-two percent
of subjects had no history of recurrent headaches. In 8 of 11 subjects
(73%) who had a documented family history, recurrent headaches
occurred in siblings, parents, or children. The number of episodes
ranged from 1 to 500 and was 10 or more in 69% of subjects. In
the majority (77%) the episodes lasted either several minutes
or 15 to 60 minutes. The episodes lasted seconds in one case (4%), and
their duration varied from seconds to hours in another (4%).
Positive visual phenomena such as bright images ("bright lines,"
"flashes of light," "silver lightning," "sunburst,"
"disco lights"), colors ("zigzag rainbow colors," "colored
lights"), and movement of images ("dancing," "jumping,"
"wavy or wiggling lines") were often reported (77%). Frequent
descriptions of visual images included zigzag, wavy, or wiggling lines
and wavy blurring of vision "as if heat is rising from the
pavement" (38%). Exclusively negative visual phenomena were
described in only one subject who experienced for more than 20 years
"monthly episodes of marked decrease in peripheral vision
either to the left or right lasting 5 minutes and not associated with
any other symptoms." The most common pattern of visual loss was a
hemianopia (27%). Neurological symptoms (other than migrainous visual
symptoms) were associated with the episodes in 19% of subjects and
included diplopia, dizziness, numbness, paresthesias, tinnitus, and
aphasia. The migrainous nature of these symptoms was suggested by their
occurrence in association with typical visual migrainous phenomena and
the progression from one neurological symptom to the other in a
stereotyped fashion in multiple episodes. Associated nonneurological
symptoms included nausea in four subjects and eye pain in one.
View this table:
[in a new window]
Table 3. Characteristics of 26 Subjects Who Reported Episodes
Corresponding to Migrainous Visual
Symptoms ). Among the cohort members with TIAs
who developed a stroke, two thirds developed their stroke within 6
months of TIA onset.
View this table:
[in a new window]
Table 4. Stroke Rate in Subjects With Migrainous Visual
Symptoms, Subjects With TIAs, and Subjects Without TIAs or Migrainous
Visual Symptoms
Discussion
Top
Abstract
Introduction
Subjects and Methods
Results
Discussion
References
These data indicate that visual migrainous phenomena are not rare
since they occur in 1.33% of women and in 1.08% of men in a general
population sample. Previous studies have used various methodologies to
estimate the prevalence of these phenomena. Rasmussen et
al17 conducted a general health survey with a
focus on headache disorders in Denmark using the IHS criteria. Among
740 persons, lifetime prevalence of migraine was 16% (119/740), and
that of migraine aura without headache was 0.95% (7/740). Details on
the clinical features and personal characteristics of these subjects
were not reported. Since subjects with visual migrainous phenomena may
not seek medical attention, studies based on medical diagnosis are
likely to underestimate their prevalence. In one such study, Stang et
al18 used a modification of the IHS criteria and
identified 629 residents of Olmsted County, Minnesota, who met the
criteria of migraine. Aura without headache was identified in 6.3% of
the cohort (11% of men and 4.3% of women), and the mean age of
presentation was 43.2 years (SD, 15.6). In clinical series,
Alvarez19 found in 618 cases of migraine with
aura that 12% of men and 0.7% of women experienced episodes of
migrainous visual symptoms without headache. Selby and
Lance20 reported migraine accompaniments in 6
(1.2%) of 500 subjects. Two of them had suffered typical migraine in
earlier years.
This study was based on a general population sample and thus
minimized the bias of case selection encountered in clinical series.
Cohort members were systematically queried every 2 years on the
occurrence of visual symptoms during an 18-year period, ensuring that
as many index cases as possible were identified. Because of the
prospective study design, stroke risk in subjects with migraine
accompaniments could be directly compared with cases with TIA and cases
without migraine accompaniments within the same cohort. Because of the
nature of the study we had to rely on documentation of symptoms in the
medical record, which were obtained before the publication of the
IHS criteria.12 Therefore, fulfillment of the
criteria of migraine with aura or migraine aura without headache could
not be verified in the majority of subjects. Nevertheless, we believe
that the migrainous character of the visual episodes was convincing and
consistent with the diagnosis of migraine. Finally, the
Framingham population is predominantly white, and it would be difficult
to justify generalizing the results of this study to other racial
groups.
Episodes of migrainous visual symptoms in mid or late life are not
rare and occur in 1.33% of women and 1.08% of men. These episodes may
occur for the first time after age 50 years, in the absence of
headache, and a history of recurrent headaches may not be present.
Diagnosis is largely based on the clinical features of the episodes;
however, noninvasive tests (carotid duplex, transcranial
Doppler, MR angiography) may occasionally be indicated to exclude
vascular disease in cases with an atypical presentation.
Migrainous visual symptoms appear not to be associated with an
increased risk of stroke, and invasive diagnostic
procedures or risky therapeutic measures are generally not
indicated.
Acknowledgments
From the Framingham Heart Study of the National Heart, Lung, and
Blood Institute (supported by NIH/NHLBI contract N01-HC-38038) and
grants 2-RO1-NS-17950-16 (National Institute of Neurological Disorders
and Stroke). The authors wish to acknowledge Jennifer L. Burg for
assistance in the collection of data and Donna L. Copenhafer, PhD, for
assistance in data processing.
References
Top
Abstract
Introduction
Subjects and Methods
Results
Discussion
References
This article has been cited by other articles:
 |
|
 |
|
  P. E. Stang, A. P. Carson, K. M. Rose, J. Mo, S. A. Ephross, E. Shahar, and M. Szklo Headache, cerebrovascular symptoms, and stroke: The Atherosclerosis Risk in Communities Study Neurology, May 10, 2005; 64(9): 1573 - 1577. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Fornadley Using Vision Changes to Differentiate Sinonasal Headache From Migraine Arch Otolaryngol Head Neck Surg, February 1, 2000; 126(2): 231 - 233. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||