Bilateral caloric hyporesponsiveness has been reported in up to 11 % [1, 10, 14, 15] and an isolated directional preponderance of caloric responses in about 10 % of patients with VM [1, 4, 16]. Interestingly, patients with VM are four times more likely to have an emetic response to caloric stimulation than patients with a vestibular disorder coexisting with migraine [16].

A pathological head impulse test has been reported in up to 26 % of patients with VM [13], but in our experience, it occurs only exceptionally [10].

Rotatory chair testing revealed an isolated directional preponderance in about 20 % of patients [3, 4]. Some authors reported a reduced gain of the horizontal vestibulo-ocular reflex during rotatory chair testing [17, 18], but this finding was confirmed in only 1 % of patients in a large case series [3].

Assessment of cervical and ocular vestibular-evoked myogenic potentials (cVEMPs and oVOMPs) has yielded either unilaterally or bilaterally reduced amplitudes in about two-thirds of patients with VM indicating saccular and utricular dysfunction [19, 20]. Another study found absent cVEMPs in 43 % of patients with VM [21]. Cervical recorded VEMPs were also absent in 35 % of 20 patients with basilar-type migraine, most of them experiencing vertigo [22]. The latencies of the response were only rarely prolonged in patients with VM [23], being normal in most studies [19–21]. VEMPs do not seem to be helpful for the differentiation of VM from Menière’s disease, where similar results can be found [19]. One study elicited VEMPs applying tone bursts of various frequencies and concluded that this method may help to separate VM from Menière’s disease, but these results await replication [24].

A large case series of patients with VM yielded normal results of posturography in 74 % of patient [3]. Group analysis of posturography demonstrated excessive reliance on somatosensory cues [6] or on visual cues [8, 9] in patients with VM as compared to controls.

Saccadic pursuit has been reported in 3 % [3] to about 57 % [11] of patients with VM. No other ocular motor finding has been reported with such a wide variance in patients with VM. Most authors found that saccadic pursuit occurs in about 10–20 % of patients with VM in the interval (Table 7.1). Assessment of smooth pursuit is problematic as it relies on attention and cooperation of the patient. Furthermore, the vast majority of studies assessed smooth pursuit clinically without eye movement recording. Two case series that described saccadic horizontal smooth pursuit in about half of patients with VM found impaired fixation suppression of the vestibulo-ocular reflex (VOR) in only 3 % of these patients [4, 11]. These are conflicting findings as cancelation of the VOR is typically impaired when smooth pursuit is saccadic [10].

Spontaneous nystagmus is rare in the interval with a prevalence of well below 10 % in most case series (Table 7.1). In contrast, positional nystagmus of a central type is not uncommon and has been described in about 10–20 % of patients (Table 7.1). Gaze-evoked nystagmus occurred in less than 5 % of patients in several case series [6, 9, 10, 13, 25], and only Dietrich and Brandt observed gaze-evoked nystagmus in a large proportion of patients (27 %) with VM [4]. Head-shaking nystagmus has been described in 15–50 % of patients [10, 12, 13, 25] and can be horizontal or downbeating [25]. Vibration-induced nystagmus typically indicating peripheral vestibular hypofunction [26] has been observed in 32 % of patients with VM [12].

It is important to notice that these clinical and laboratory findings are not specific to patients with vestibular migraine but can also be found in migraine patients without a history of vestibular symptoms. A unilateral canal paresis has been described in up to 35 % of migraine patients without vertigo [9, 13, 27]. Clinical examination yielded head-shaking nystagmus in 9–25 % of migraine patients without vertigo [13, 25]. Likewise, central positional nystagmus has been described in patients with migraine without a history of vestibular symptoms [28]. While a high frequency of pathological oculographic findings has been reported by several authors in migraine [5, 9, 27–29], other authors failed to find significant abnormalities [30, 31]. Several studies examined the prevalence of vestibular dysfunction in patients with vestibular migraine as compared to migraine patients without vertigo. In two studies. the prevalence of peripheral and central vestibular dysfunction did not differ between both groups [5, 9], whereas another study reported a higher prevalence of central and peripheral vestibular dysfunction in patients with VM (70 %) than in migraine patients (38 %) [13]. In particular, unilateral caloric hyporesponsiveness has been found with similar frequency in migraine patients with and without a history of vestibular symptoms [9, 13]. Only saccadic pursuit seems to be more frequent in VM as compared to migraine without vertigo [9, 13].

Two studies examined the evolution of interictal vestibular and ocular motor dysfunction in patients with VM over time. In a group of 61 patients with VM, the prevalence of at least one ocular motor abnormality increased from 15 % at initial presentation to 41 % after a median follow-up time of 9 years [10]. The most frequent abnormalities were positional nystagmus and head-shaking nystagmus (Table 7.2). Definite central-type positional nystagmus was present at follow-up in 18 % of patients. Another 8-year-long observational study with 30 patients found that the prevalence of central ocular motor deficits increased from 20 to 63 % in VM [11]. In this study, the most common finding was saccadic pursuit.

In general, signs of central ocular motor and vestibular dysfunction remain subtle throughout the course and do not worsen over the years [10, 11]. Interestingly, interictal ocular motor abnormalities may show some variation over time [4, 10], and in some patients ocular motor dysfunction may even turn to normal at follow-up [10]. Thus, ocular motor abnormalities observed in the symptom-free interval may partly reflect delayed recovery of vestibular dysfunction after an acute vertigo attack.

Audiometry revealed sensorineural hearing loss not attributable to any cause in up to 20 % of patients [15]. A review on audiometric findings in vestibular migraine summarized results of nine studies and found an average prevalence of unexplained hearing loss of 7.5 % [32]. Thus, hearing loss is rather unusual, and low-frequency, progressive, or fluctuating hearing loss, typical for Menière’s disease, is a rare finding in vestibular migraine. In a case series of 61 patients with VM, 18 % of patients had developed mild bilateral sensorineural hearing loss with a downsloping pattern involving also the low-frequency range after a median follow-up time of 9 years after initial presentation [10].