Late-life migraine accompaniments: A narrative review

Epidemiology

The one-year prevalence of migraine declines in the aging population (7,8,11,12,17). However, migraine aura without headache can occur at any age and is common in elderly patients (13,15,16). The largest case series of late-life migraine accompaniments were reported by Fisher (120 cases in 1980 and 85 cases in 1986, 205 cases in total). From the first series of 120 cases, 35 cases presented with transient episodes that resembled the neurological accompaniments of migraine. Headache occurred in 50% of cases (15). Analyses of 85 additional cases showed that they affect 23% in patients aged 40–49 years, 40% in 50–59 years, 20% in 60–69 years and 16% in 70 years and over. Headaches were associated with only 40% of these episodes and 65% of patients had a history of recurrent headaches in the past (16).

In the population-based study from the Framingham cohort conducted with 2110 individuals, 186 (8.8%) reported first-ever episodes of sudden visual defect. Fourteen percent of those (26 of 186 people) reported visual symptoms that corresponded to visual aura in migraine. The age of onset was greater than 50 years in 77% (20 of 26 patients). Mean age of onset was 56.2 years. Fifty-eight percent (15 of 26 patients) of those with visual symptoms never had headache. Forty-two percent had no history of recurrent headaches. The overall prevalence of migrainous visual symptoms in the general population was 1.23% (1.33% in women and 1.08% in men). The overall prevalence of migrainous visual symptoms without headache was 0.71% (0.70% in women and 0.72% in men) (13).

A one-year prospective study (Marseille’s registry from a tertiary headache clinic in France), enrolled 3773 migraine patients. Migraine with aura was reported in 26% (978 of 3773 patients). Of these, 6% (57 of 978 patients) were aged 50 years and over. About 30% of patients (18 of 57 patients) experienced migraine aura for the first time after age 50. Prevalence of first-time migraine aura after age 50 was 1.8%. Typical aura without headache was reported by 61% (11 of 18 patients). Occurrence of aura exclusively without headache was reported by 17% (three of 18 patients) (18).

A cross-sectional study identified individuals who presented at least one episode of unexplained transient focal neurological symptoms suggestive of cerebral dysfunction, but no history of migraine headache among physicians (233) and inpatients (690) of an academic hospital. A migraine screening questionnaire, electroencephalography (EEG), and brain magnetic resonance imaging (MRI) were performed. Cortical hyperexcitability, assessed by occipital transcranial magnetic stimulation, was used as a marker of possible migraine aura without headache. Migraine with aura was reported in 20%; transient neurological symptoms unrelated to headache that clinically resemble migrainous auras in 9%. Abnormal occipital transcranial magnetic stimulation was observed in 75% of cases (none in the control group), supporting the diagnosis of migraine aura without headache (19).

A study of typical aura without headache was conducted in ophthalmology clinics. This study included 1063 outpatients using a seven-item self-reporting questionnaire (ID Migraine screener, Japanese version) with six additional questions. Visual and/or sensory symptoms that corresponded with migraine aura were diagnosed in 3.2% (35 of 1063 patients) with 1.1% in males and 2.1% in females (male:female ratio = 1:1.19). The median age of patients of typical aura without headache was in the fourth decade (median age 47 years). The age of patients who were diagnosed with typical aura without headache showed two levels of age distribution: the second to third and sixth decades (20–39 years and 60–69 years) (20). Another survey study from ophthalmology clinics, conducted in 1000 primary eye care patients, showed visual phenomena that are consistent with typical aura were reported in 6.5% of cases (21). Typical aura without headache appears to be reported more frequently in ophthalmology than in neurology clinics (21,22).

In a prospective population-based study from Italy (the Bruneck study), 574 patients between 55 and 94 years underwent neurological examinations, laboratory tests, and headache interviews. The prevalence of migraine and female predominance decreased in the elderly. Men had more migraine attacks than women. Lifetime prevalence for typical aura without headache was 0.9% (men 1.6%, women 0.3%) and one-year prevalence was 0.3% (men 0.8%, women less than 0.3%) (23).

A population-based study from Denmark was conducted in 740 patients. Extensive descriptions of the individuals’ headache history were taken, and general physical and neurological examinations were performed. Headache disorders were classified according to the IHS diagnostic criteria. The lifetime prevalence of migraine was 16% (119 of 740 participants). Prevalence of migraine aura without headache was reported in 0.95% (seven of 740 participants) (24).

Population-based research from Olmsted County, Minnesota, US, showed 629 of 6476 patients presenting to the hospital were newly diagnosed migraine headaches according to the IHS’s 1988 criteria. Incidence of aura symptoms without migraine headache was 6.3% (11% in men and 4.3% in women), and mean age at onset was 43.2 years (SD 15.6 years) (25).

A prospective study of migraine with aura was conducted in a headache clinic population. Sixty-four patients who had migraine with aura, according to the IHS classification in 1988, were included in the study. Migraine aura without headache was reported in 6% of these patients. After a follow-up period of six to 15 months, the prevalence of migraine aura without headache was 29%. In addition, attacks of migraine aura without headache were significantly more frequent in men than women (26).

Observational studies from 618 migrainous scotoma patients found episodes of migraine aura without headache in 12% of men and 0.7% of women (27). Selby and Lance analyzed 500 cases of migraine. Migraine accompaniments were reported in 1.2% (six of 500 participants) (28). Ziegler and Hassanein observed specific headache phenomena in 392 patients who attended a headache clinic and found migraine with aura without headache in 44% of migraine aura patients (29).

Clinical features

Migraine tends to decrease with age, but it is still a relatively common complaint in the elderly (17,30–33). Clinical characteristics of migraine change over time with age (34–36). Migraine attacks were less typical in the elderly group. Premonitory symptoms including anorexia, paleness, and dry mouth were more common in the elderly (37). Aura and associated symptoms of photophobia, phonophobia, and dizziness were less common in the elderly. Neck and occipital pain were more frequent than temporal pain. Patients over 50 years had milder headaches, greater ability to function during headache, and good response to acute medication (35).

Migraine aura may be experienced for the first time in the middle aged or in the elderly (38–41). Late-life migraine accompaniments may occur for the first time in 33%–77% of migraine with aura after 50 years of age (13,18). Unlike migraine in general, migrainous visual aura symptoms did not decrease in advanced age. Aura often continued in the absence of headache (15,16,36–38,40,42). About half of migraine with aura patients had some aura without headache (39,41). Stereotypic aura episodes were reported in 65% (13). Visual auras were most frequent (92%–100%), followed by sensory aura (28%–31%), aphasic aura (18%–33%), and motor aura (2%–6%) (9,11,15,16,18,26,39,43). Various combinations of aura symptoms occurred, whereas sensory, motor, and aphasic symptoms always occurred accompanied by visual aura (15,16,18).

The symptoms of late-life migrainous accompaniments were described by Fisher more than 30 years ago (15,16). He classified accompaniments into nine categories: 1. visual symptoms (blindness, homonymous hemianopia and blurred vision, difficulty focusing), 2. visual and paresthesias, 3. visual and speech disturbances (aphasia, and/or dysarthria), 4. visual and brain stem symptoms, 5. visual, paresthesias, and speech disturbances, 6. visual, paresthesias, speech disturbances and weakness, 7. only non-visual accompaniments (paresthesias, speech disturbance, weakness, etc.), 8. recurrence of old stroke deficits, and 9. miscellaneous. Classification of migraine accompaniments is showed in Table 2.

OPEN IN VIEWER

Table 2.

Classification of migrainous accompaniments with frequency of distribution Adapted from Fisher (15,16).

Classification of migrainous accompaniments	Number		Percentage
1980 (15)	1986 (16)
Visual (excluding scintillating scotoma)			23%
Blindness	14	3
Homonymous hemianopia	5	7
Blurred vision, difficulty focusing	6	9
Visual and paresthesias	18	6	12%
Visual and speech disturbance (dysarthria or aphasia)	7	2	4%
Visual, paresthesias and speech disturbance	7	3	10%
Visual, paresthesias, speech disturbance and paresis	7	20	5%
Visual and brain stem symptoms	14	3	14%
No visual accompaniments (only paresthesias, speech disturbance, paresis, etc.)	25	32	30%
Total	103	85	100%

Visual auras

Visual aura is the most common aura and does not always occur at the same time with other accompaniments (13). Typical visual aura starts as flickering, uncolored, unilateral zigzag lines, more often beginning centrally in one or both visual fields, and it gradually builds up and progresses toward the periphery. Most visual auras last 15 to 30 minutes, with an average duration of 15 minutes (9,15,16,39,41–43). The most common visual symptoms are scintillation scotomas (fortification spectra, teichopsia), often bright and shimmering. Positive visual symptoms include bright images (bright lines, small bright dots, flashes light, white spots, disco lights), colored lights (zigzag rainbow colors), and movement of images (wavy lines, dancing, jumping) and occur in up to 90% of patients (9,13,15,16,18,41,43,44). Negative visual phenomena include scotomas (blind spot or partial loss of sight, often crescent shaped), blurred/foggy vision, and black dots/spots occur in up to 50% of patients (9,39,41). The presence of both positive and negative features is the hallmark of visual aura. Visual perception disturbances including “Alice in Wonderland,” metamorphosia, micropsia, macropsia, tunnel vision, dyschromatopsia, hypo/achromatopsia, bright multi-colors, visual agnosia, prosopoagnosia, and other visual distortions have also been reported (9,15,16,43–48).

Lashley, in 1941, provided a quantitative detailed record of his own migrainous visual auras (49). He plotted their nature and temporal spread. Common characteristics of his auras included the following: Aura always started at or near the center and gradually spread peripherally of visual fixation and were almost symmetrical in both eyes. They never imposed on the midline; some scotomas were exclusively negative, others scintillated at the edge, which appeared serrated or zigzagged. As the scotoma moved laterally, it increased in size and in the rate of drift and involved both the upper and lower quadrants of the right or left hemifield of vision; occasionally only the lower or upper quadrant was affected. The size and density of scotomas varied, and complete hemianopsia was observed only once in more than 100 descriptions of visual aura. Close examination of the scintillations revealed a series of parallel zigzag lines. There was a tendency for the topography of the advancing front to remain relatively constant in both size and shape as the fortification drifted laterally. The duration of the scotoma after the scintillations subsided varied from approximately five to 30 minutes. He calculated the wave propagated across the visual cortex at a rate of about 3 mm per minute, resembling the speed of cortical spreading depression (CSD) discovered by Leão in 1944 (50).

Fisher described his own 27-year experience of 41 episodes of typical migrainous scintillation zigzags without headache in late life (between the ages of 59 and 85). Episodes started without warning with central scotomas near the fovea. After 40 seconds to five minutes, a scintillating zigzag line in one hemivisual field followed. Two phases (early and late) were described. In the early phase, a scintillating line consisted of an inner zigzag black line and outer side bright zone. After five to 30 minutes, the scintillating line gradually expanded into a scintillating arc of increasing length and migrated toward the periphery of the involved visual field. Arms of each zigzag unit became reduplicated into four or five black and bright lines that swept repetitively from inward to outward. In the late phase the scintillating zigzag lines moved more temporally, more than 30 meridian angles from the fovea, and were less clearly seen. Zigzag lines then faded and disappeared (51).

A 71-year-old man had more than 1000 attacks of migraine aura without headache over 18 years. Most auras originated centrally (within 10-degree eccentricity) and propagated predominantly in the lower nasal field bilaterally (69%–77% of all auras). They spread to the upper temporal fields. Some auras propagated from peripheral to central visual field, typically following the same path as those propagating in the opposite direction. The speed of cortical propagation was estimated to range from 2 to 3 mm/min. Visual perception was characterized by a colorless curvilinear band of flickering (10–15 Hz) brilliant-and-black gridlines (fortification spectra), followed by an advancing area of impaired vision (scotoma). Around 20% of attacks showed limited propagation and had shorter duration than usual auras (52).

Sensory auras

Paresthesias are the second most common migrainous accompaniments, occurring in up to 30% of patients who have migraine with aura (15,16,53). Most patients who have sensory aura also have visual aura (28,39). Sensory auras include positive symptoms (tingling, paresthesia, pins and needles sensation) and negative symptoms (numbness). Cheiro-oral (hand and face) distributions occur in 97% of patients with sensory aura; the body and legs are rarely affected (54). Typical cheiro-oral auras were predominantly unilateral, starting in the hand, slowly spreading up to involve the forearm, and then migrating to the ipsilateral face, lips, chin, and tongue. Lower extremities are occasionally involved. Sensory auras most often involved hands (96%) and face (67%), whereas legs (24%) and torso (18%) were much less commonly affected (39,53,54). Sensory symptoms were bilateral in 50% of patients (44). They last for 20 to 30 minutes and then clear in the reverse order (areas affected first, clear last) in 82% (39,55).

Dysphasic auras (speech disturbances)

Dysphasic auras include aphasia, dysarthria, impaired production of language, and impaired comprehension of language. Dysphasia is characterized by difficulty finding words, difficulty naming objects, and difficulty producing or understanding language. Most patients with impaired production of language had paraphasia (11,15,16,39). Language aura presented as paraphasic errors in 76%, other production problems in 72%, and impaired comprehension in 38% of cases (39). In Fisher’s series, dysphasic auras were reported in 22%. Dysarthria was experienced by 46% of patients with aphasic aura (11). Median duration of aphasic aura was 30 minutes (11,15,16).

Motor auras

The ICHD-3 beta criteria do not include motor weakness in the typical aura symptoms; migraineurs who have motor weakness are classified as hemiplegic migraine (10). But weakness or paralysis can occur in late-life migraine patients and it is often associated with others accompaniments, most commonly visual and sensory auras (15,16). Typical motor auras were characterized by unilateral location and frequently affected hands and arms in up to 90% (39,43). Motor auras were reported in only 5% of Fisher’s series and 6% of patients who have migraine aura (15,16,39). In some patients, profound numbness may be misinterpreted as weakness but a unilateral facial droop can confirm true weakness.

Non-familial migraine with unilateral motor symptoms (MUMS) was described by Young et al. Of the 24 patients with MUMS, all patients reported weakness of the arm and hand, 17% reported facial weakness, and 8% reported weakness of the leg. A “march” of weakness was reported in 79% of patients, which was rostrocaudal in 95% and caudorostral in 5%. In total, 58% had persistent weakness without headache or headache exacerbation. Sensory symptoms were reported in 92% of patients with MUMS, with 64% reporting a march of sensory symptoms (56).

Table 3 presents the main typical features of late-life migrainous accompaniments.

OPEN IN VIEWER

Table 3.

Typical clinical features of late life migrainous accompaniments (15,16,51,52).

Clinical features of late life migrainous accompaniments
1) Presence of visual symptoms, scintillating scotomas (as an important clue).
2) Gradual built-up expansion and migration of scintillation.
3) March of paresthesias.
4) Series of progression from one accompaniment to another and from one body part to another in those with sensory or motor symptoms.
5) Occurrence of two or more similar spells.
6) Headache in association with spell.
7) Duration of episode last 15 to 25 minutes.
8) Occurrences of flurry accompaniments in 50 to 60 years of age.
9) Generally benign course.
10) Exclusion of cerebral thrombosis, embolism, dissection, subclavian steal, epilepsy, thrombocythemia, polycythemia, hyperviscosity syndromes, and lupus anticoagulant.
11) Normal angiography.
12) Recurrence of essentially identical spells over 5 to 10 years.

Diagnosis and differential diagnosis

Late-life migraine accompaniments while benign must be differentiated from secondary causes. Transient neurological disturbances in migraine can mimic other serious conditions (57–62). They can be difficult to distinguish from TIAs (60,63,64) and seizures (65–72). Other potential differential diagnosis which can produce symptoms similar to migraine aura are subarachnoid hemorrhage (73–78), cerebral amyloid angiopathy (79–84), arteriovenous malformation (AVM) (85–88), dural arteriovenous fistula (dural AVF) (89), patent foramen ovale (90–97), posterior circulation embolism (98), internal carotid artery dissection (99–101), vertebral artery dissection (102–104), carotid artery stenosis (105,106), Moyamoya disease (107), and brain tumors (108–111). Rare conditions including mitochondrial diseases (112–114), atrial myxoma (115), cerebral vasculitis (116), systemic lupus erythematosus (117–119), antiphospholipid antibody syndrome (120), Sneddon’s syndrome (121–123), essential thrombocythemia (124–126), polycythemia vera (127,128), hereditary hemorrhagic telangiectasia (Osler-Weber-Rendu disease) (129,130), leptomeningeal angiomatosis (Sturge-Weber syndrome) (131,132), Hashimoto’s encephalopathy (133) can also mimic migraine aura.

Structural lesions may share a common pathophysiology, such as increased susceptibility to generate CSD, which may produce similar clinical features of migrainous auras. Migrainous auras often start insidiously (within minutes), in contrast to vascular events or epileptic seizures that develop suddenly (within seconds). Clinical clues, which should raise suspicion for secondary causes, include short or long duration (<5 minutes or >60 minutes), negative phenomena (loss of vision or numbness), association with weakness, increased frequency or change in pattern of aura (58,59). Diagnosis of late-life migraine accompaniments should be made by exclusion with appropriate investigations (16,134).

TIAs

Late-life migraine accompaniments and TIAs can be differentiated by history and clinical features (Table 4). A history of previous migraine attacks with aura may be helpful, especially if prior aura features are similar to those experienced later in life. Migrainous auras are of gradual onset, evolve and expand over time, have sequential accompaniments from one to another modality (visual, sensory, speech, motor), from one to another body part (e.g. hand then face), and repetitive stereotypic episodes (60,63,64) (Figure 1). OPEN IN VIEWER

OPEN IN VIEWER

Table 4.

Distinguishing features of late-life migraine accompaniments versus TIAs (9,15,16,51,52,134–137).

	Migraine aura	TIAs
Visual	- Positive visual phenomena (bright)	- Negative symptoms (loss of vision)
	- Involvement of both visual fields	- Unilateral
	- Slowly moving across visual field	- Static
- Average duration 15 to 60 minutes	- Average duration 3 to 10 minutes
Sensory	- Gradual build-up/evolution	- Abrupt onset
	- Positive phenomena (paresthesia)	- Negative symptoms (numbness)
	- Cheiro-oral distribution (hand and face)	- Unilateral paresis
	- Repetitive attacks of identical nature	- Variable attacks
	- Sequential progression from one modality to another (visual, sensory, speech)	- Simultaneous appearance of symptom modalities (e.g. motor and sensory) and body parts (face and arm at same time)
	- Sequential progression from one body part to another	- Symptoms appear and disappear simultaneously
	- Area involved first clear last	- Flurry of stereotypic attacks not common
	- Flurry of attacks in midlife common	- Average duration 5 to 10 minutes
- Average duration 20 to 30 minutes

TIAs: transient ischemic attacks.

Visual symptoms in migraine are slow in onset, evolving in one or both visual fields (homonymous) and enlarging, moving across visual fields lasting several minutes in almost all cases. Visual abnormalities are often bright and shimmering (positive phenomenon). Average duration usually lasts for 15 to 30 minutes. In TIAs, visual defects are invariably abrupt in onset and comprise negative visual symptoms (loss of vision, dark, dim) in one eye and a static course. Duration of attacks is often shorter, averaging 3 to 10 minutes (135,136).

Sensory symptoms are typically positive phenomenon (paresthesias) in migraine. Symptoms begin in the distal part of the extremities and spread slowly up or down, most commonly starting in the fingers, then slowly marching up to the hand or forearm, and may also involve the ipsilateral face, lip, tongue, and shoulder. Paresthesias tend to last 20 to 30 minutes and then clears in reverse order (areas that are affected first, clear last). The sensory symptoms in TIAs are a sense of numbness or sensory loss, Sensory symptoms occur suddenly, lasting for five to 10 minutes and resolving in the same order they develop. The tingling sensation moves across the face to the perioral area and sometimes involves half of the tongue. Symptoms involving the buccal mucosa or tongue are rarely seen in patients with TIAs (134,137–139).

Motor symptoms are usually mild in migraine. Motor aura is always unilateral, usually affecting the hand and arm, and almost always associated with gradually developing sensory symptoms on the affected parts (12,39). Unlike sensory auras, areas where the motor deficit first appear, clear first (134). Diffusion-weighted imaging (DWI) sequence brain MRI can help identify acute ischemic lesion in approximately 30% of TIA patients (140–142). Cerebral MRI with susceptibility-weighted imaging (SWI) has diagnostic value to differentiate hemiplegic migraine from other causes of neurological deficit and may be considered in some patients who have acute motor weakness that is clinically consistent with hemiplegic migraine, especially in the first episode (143).

Subarachnoid hemorrhage (cSAH)

Focal cortical, non-aneurysmal, non-traumatic cSAH in the elderly can cause transient focal neurological episodes (TFNEs), including positive visual phenomenon, spreading pattern paresthesia, negative symptoms of numbness, stereotyped focal weakness, and speech difficulties (72–84). They mimic auras of late-life migraine accompaniments, TIAs, or focal seizures. The syndromes of “crescendo transient aura attacks” were described in patients with focal subarachnoid hemorrhage as transient, sensory, stereotyped and recurrent, slow to evolve, and migratory in nature (74).

TFNEs are lately recognized as the one clinical manifestation of cerebral amyloid angiopathy (CAA), sometimes termed “amyloid spell” (81). A multicenter cohort study found TFNEs in 14% of CAA patients who presented with predominantly positive symptoms (aura-like) in 52% and negative symptoms (TIA-like) in 48% (82). The most common positive symptom was gradual spread of transient paresthesias affecting the mouth and hand that lasted less than 30 minutes in most cases. Transient zigzags, flickering or flashing lights, monocular blurring of vision, visual loss, and focal weakness were also reported (79–84).

Clinical features of TFNEs and radiological syndrome of focal SAH are strongly associated with CAA (82). MRI with T2*-weighted gradient-recalled echo (T2*-GRE) or SWI sequences revealed that TFNEs may be caused by CAA-related hemorrhage (82,84,144–146). In patients older than 60 years old with focal cortical SAH, the most common cause is CAA (75,81,82). A recent study found that cortical superficial siderosis in CAA reflected previous hemorrhage in the superficial layers of the cerebral cortex or in the subarachnoid space, and suggested that cortical superficial siderosis should be considered a marker of CAA-related hemorrhage (144–146).

There is a very high, early risk of symptomatic lobar intracerebral hemorrhage (37.5% at two months) in patients with CAA-related TFNEs (80,82,147,148). The pathophysiology of TFNEs associated with CAA remains unknown. However, CSD, induced by subarachnoid hemorrhage, has been demonstrated both in experimental and human studies (149–151).

Investigations

Primary headaches are still predominant in the elderly but secondary headaches increase in this population (58,137–139,152,153). Underlying life-threatening causes should be identified by appropriate laboratory tests and neuroimaging studies (154). Structural cortical lesions may increase susceptibility to CSD, which produces similar transient phenomenon resembling migrainous auras (59). Workup to rule out secondary causes such as ischemic conditions (TIAs, stroke, SAH, dissection), seizures, vascular abnormalities (AVM, dural AVF), and neoplasms may need to be performed before a diagnosis of late-life migrainous accompaniments can be made. If imaging is needed, cerebral MRI with or without contrast, including DWI, T2*GRE, or SWI sequences, should be performed; MRI has higher sensitivity for detecting these and other lesions than computerized tomography (74,75,155–157). However, several studies have demonstrated the association between migraineurs and brain MRI abnormalities, including white matter lesions and subclinical posterior circulation territory infarct-like lesions, especially in migraine with aura (158–164). Magnetic resonance angiography or vascular ultrasound should be performed when carotid artery dissection, vertebral artery dissection, carotid artery occlusion, or vascular malformations are suspected. EEG should be performed to rule out seizure disorders when differentiation based on clinical features is uncertain (71,72,165).

Treatment

Late-life migraine accompaniments are benign (13,15,16,134). Treatment of migraine in elderly patients requires special consideration of comorbid medical illnesses (166,167). Elderly patients are more susceptible to adverse effects of medications. Non-pharmacologic approaches including lifestyle modifications, exercise, biofeedback, and cognitive behavioral therapy should be considered (168–170). Medications should be started at a low dose and titrated slowly, closely monitoring clinical response and adverse events.

Studies of late-life migraine accompaniment treatments were very limited. Amitriptyline ameliorating aura symptoms was reported in only one report (34). There is no specific treatment for late-life migraine accompaniments, although standard migraine prophylaxis is generally accepted. Migraine accompaniments usually resolve spontaneously in less than one hour without treatment. However, for patients with aura status, prolonged or persistent aura treatment may be considered.