Clinical characteristics of Alice in Wonderland syndrome in a cohort with vestibular migraine

Discussion

AIWS was described by 14% of the patients diagnosed with VM in our clinic, close to the estimated 15% prevalence rate of AIWS in migraine patients.¹ AIWS disturbances began with the onset of VM in all our patients. The mean age at onset was 42.1 years, which is close to the mean age of 40 years for adult-onset AIWS,¹ but higher than the mean age at onset of VM reported by others (35–38 years).^11,12

Similar to other studies,^1,5,13 most of our patients experienced visual distortions. Of the 42 visual distortions previously described in AIWS,¹ those experienced by our cohort include illusory splitting (patient 1), xanthopsia (patient 9), and enhanced stereoscopic vision (patient 16). Four rare and very unusual visual distortions were described. Patient 10 described the “dolly zoom” effect, a distortion of visual perspective produced by zooming the camera lens onto an object while simultaneously moving the camera farther from that object in such a way as to maintain the object's size in the frame throughout, causing the background to change in size relative to the object of interest. The dolly zoom effect was first conceived of and used in Alfred Hitchcock's film Vertigo.¹⁴ Patient 8 described “underwater vision” and patient 13 reported “frosted-glass vision.” Another curious symptom was patient 17's closed-eye visual hallucinations, hallucinations that only occur when a patient's eyes are closed and immediately disappear with eye opening. Closed-eye visual hallucinations have previously been reported following general anesthesia, local anesthesia with lidocaine, and atropine poisoning.^15,–,19 While hallucinations are not strictly a sensory misperception, these closed-eye visual hallucinations may be considered an AIWS symptom in view of its link to the patient’s VM attacks, and because hallucinations (e.g., zoopsia) have been considered manifestations of AIWS in previous publications.^3,8

An unusual finding in our cohort is the high proportion (41%) with extrapersonal misperceptions (4 with out-of-body experiences, 2 with derealization, 1 with depersonalization) compared to other migraine-related AIWS studies.^1,–,4 In fact, while the most common type of AIWS misperceptions in our cohort was visual in nature, the most frequent AIWS symptom described was an out-of-body experience (OBE). It is likely that the higher proportion of extrapersonal misperceptions in our cohort may be related to the vestibular symptoms accompanying the migraines (and not just headaches), in view of the fascinating relationship between vestibular symptoms and such phenomena. Derealization/depersonalization symptoms are not infrequent in peripheral vestibular disorders.^20,21 Patients experiencing OBE frequently report associated vestibular symptoms, regardless of the etiology of the OBE²²; in fact, OBE have been observed in both central and peripheral vestibular disorders.²³ Since peripheral vestibular disorders can provoke extrapersonal misperceptions, we cannot completely discount their presence in our cohort as the consequence of vertigo, rather than AIWS. However, it is important to note that the vast majority of OBE arising from peripheral vestibular disorders only occur with the first episode of dizziness/vertigo.²³ In contradistinction, the patients in our cohort experienced recurrent extrapersonal misperceptions with every VM episode, as well as between VM episodes (i.e., independent of any vertigo), suggesting a migrainous etiology instead. The temporo-parietal junction has been identified as the neurologic source of OBE^22,24,–,26; this location is not only part of the vestibular cortex (explaining the relationship between vestibular symptoms and OBE), but has also been implicated in AIWS (see later).

A greater proportion of our cohort also reported somasthetic distortions (29%) compared to other studies (approximately 10%).^5,13,27 Partial macrosomatognosia was reported by 2 (patients 12 and 14). Patient 2 experienced an unusual form of aschematia where he felt “he had no eyes.” Aschematia was also described by Todd⁸; 1 patient reported that her hands “drop off and disappear” and another described “one or other of his arms was missing.” The distortions experienced by patients 5 and 15 in our cohort are highly unusual. Patient 5 described the unusual sensation of her “brain coming out of her head” during VM attacks. Patient 15 experienced whole-body macrosomatognosia and microsomatognosia between VM episodes that were only triggered by very bright headlights from oncoming traffic when driving at night; all other reported cases of AIWS are spontaneous, and to our knowledge, not triggered by bright lights. Photophobia is an integral part of migraine, and bright lights are a well-known trigger for migraine headaches.^28,–,31 One hypothesized pathway involves a population of intrinsically photosensitive retinal ganglion cells (ipRGCs) that directly communicate with the posterior, lateral posterior, and intergeniculate thalamic nuclei (a pathway responsible for exacerbation of migraine headache by light)³²; these nuclei, in turn, are connected with the primary and secondary somatosensory cortices, which also form part of the central vestibular network.^32,33 It is possible that the somesthetic misperceptions triggered by headlight glare in patient 15 arise from photoactivation of ipRGCs that subsequently induces abnormal activity within the vestibular cortices (analogous to how light exacerbates migraine headaches), causing the distorted body spatial representation that characterized his AIWS.

In pediatric studies, AIWS symptoms are typically self-limiting and cease within weeks or months,^13,27 but in some, continue well into adulthood.⁴ All our patients reported that AIWS began at the same time as VM symptoms, and continued to experience the symptoms until the time of presentation; no adults reported childhood AIWS. We followed the treatment response of 11 after starting migraine preventive therapy. All of them reported either complete resolution or partial improvement in both VM and AIWS, except for patient 8, who experienced complete resolution of VM but not AIWS. This suggests that migraine preventive therapy can ameliorate AIWS symptoms, and could potentially be used to help those who find their AIWS symptoms distressing. Interestingly, in the 2 patients with abnormal neuro-otologic examination findings, the improvement in VM and AIWS was also accompanied by improvement in these abnormalities.

A high proportion of our cohort carried a diagnosis of anxiety disorder (43%) and depression (31%), which is unsurprising in view of the relationship of mood disorders with migraine and VM.^34,–,36 The vast majority of our patients (82%) expressed apprehension that the AIWS were symptomatic of, or would lead physicians to diagnose them with, a psychiatric illness, similar to Corbett's³⁷ observations. This underscores the need for neurologists to be aware of AIWS, and allay patient fears that the distortions of AIWS indicate mental illness.

A history of migraine without aura was reported by 24% of our cohort, consistent with other studies in VM that observe a history of migraine headaches preceding the onset of VM.³⁸ Similar to our study, a family history of migraine was present in the studies by Liu et al.¹³ and Weidenfeld and Borusiak²⁷; one patient in each of these studies also reported a family member with childhood AIWS. None of our patients reported family members with childhood AIWS in childhood but intriguingly, one described similar AIWS and VM symptoms in an adult family member (his sister).

While many publications on AIWS recognize its relationship with migraine headaches, we specifically highlight the oft-overlooked occurrence of AIWS in VM. Upon reviewing Todd's⁸ seminal case series, it is highly likely that 5 of the 6 patients he described experienced vestibular symptoms in conjunction with migraine attacks, and thus, most likely had VM. Four had vertigo; one had “giddiness with nausea,” tilt illusion (“feeling that she was either going up or down hill even though walking on a flat surface”), and imbalance (“a tendency to lurch into articles of furniture”).⁸ Two patients in Lippman's⁷ cohort also most likely had VM; one would “teeter and reel as though drunk” before, during, or after a migraine headache, while another had migraine headaches associated with “staggering, imbalance on sitting or standing.” Looking back even further, Oppenheim's 1913 patient with body image distortion experienced “spontaneous dizziness” accompanying migraine headaches.⁶ As such, it is important to emphasize how an association between AIWS and VM was observed long before VM was conceived as a distinct clinical entity.

The precise localization of AIWS remains unclear; case reports and series in AIWS have implicated widespread and different parts of the brain including the frontal, temporal, parietal, and occipital lobes.^1,–,3 This disparity may be due to the different sensory modalities affected by these distortions; for example, it is unlikely that visual distortions share an exact neurologic localization as extrapersonal misperceptions. The relationship between VM and AIWS may provide clues as to the location of the neural substrate responsible for generating this unusual disorder. The associative cortical regions responsible for generating 3D bodily self-consciousness (i.e., self-location and first-person perspective) need to continuously integrate real-time signals from exteroception (visual, auditory), somatosensation (tactile, proprioceptive), interoception (e.g., thermal, nociception), and vestibular information,³⁹ and are the most likely source of AIWS misperceptions. The neural structures involved in this multisensory integration include the posterior parietal cortex, medial superior temporal region, parietoinsular vestibular cortex, and temporo-parietal junction, which also form core components in the cortical vestibular network.^33,39,40 In fact, there is evidence that the vestibular system plays a central role in whole-body spatial representation, body part representation, self-motion perception, mental spatial representations, as well as the relationship between the external world and a global full-body representation.³⁹ While the precise pathophysiology of VM is unclear, one hypothesis is cortical spreading depression (i.e., short-lasting neuronal depolarization spreading over adjacent cortical areas, followed by a longer-lasting inhibition of neuronal activity) that involves the aforementioned multisensory integration cortical regions³⁸; it is not inconceivable that the dysfunctional neuronal activity affecting these same structures, which are critical to body schema, would cause the highly conspicuous distortions that characterize AIWS.