Sense of direction in vestibular disorders

1 Introduction

Sense of direction (SOD) can be defined as our ability to orientate ourselves within our local environment and to know in which direction objects are [48]. SOD is a complex multi-sensory process involving the integration of self-motion cues with optic flow information sensing our external environment, which is integrated at multiple levels of the neuraxis and under cognitive control [43]. The vestibular system in particular, provides sensory information on head accelerations in three-dimensional space and is known to be a key contributor to spatial orientation in the absence of visual cues [21, 41]. However, the relative importance of vestibular input to everyday SOD ability remains unclear. Many individuals are aware of having a good or poor self-perceived sense of direction ability but questionnaire tools, mainly the Santa Barbara Sense of Direction (SBSOD) scale, provide a quantitative assessment of subjective SOD skills which is representative of real-world navigational ability [25]. In addition, psychometric assessments such as the Object Perspective test [32], are useful tools to evaluate cognitive visuospatial components to SOD such as spatial updating and object orientation.

Few studies have investigated the impact of vestibular system dysfunction on SOD and, to our knowledge, no studies have compared SOD ability and the interaction of symptom effects in a large and multiple diagnoses cohort. This is important, as despite the fact that dizziness, disorientation, and vertigo symptoms commonly arise from a mismatch in visuo-vestibular processing in many disorders, mechanisms underlying differing diagnoses are diverse. Whilst structural disorders are often the consequence of physical lesions to the peripheral vestibular system, functional vestibular disorders involve more complex maladaptation of vestibular information processing in higher cortical circuitry and may be exacerbated by underlying psychological components. As such, a key question is to what extent does the form of vestibular pathology affect orientational ability? We hypothesise that dizziness and psychological symptoms, as well as overall SOD ability, likely show high variability between vestibular diagnoses. In order to assess this, we recruited a diverse cohort of patients with varying vestibular disorders, including clear peripheral vestibular diagnoses such as benign paroxysmal positional vertigo (BPPV), bilateral vestibular failure (BVF) and Meniere’s disease (MD), as well as central/functional disorders of vestibular migraine (VM) and persistent postural perceptual dizziness (PPPD) which both lack obvious physical or structural aetiology. In addition, assessing patients with BVF may help answer a more fundamental question, namely, how critical is vestibular input (severely reduced in these patients) to human SOD? Although vestibular input is essential to angular spatial orientation in the dark [34, 18], we usually refer to day-to-day SOD as an activity taking place within the visual world. Indeed, questionnaires such as the SBSOD relate to sense of orientation and direction with respect to external visual objects. Therefore, the prominent role of visual-cognitive cues in humans may mask any effects of vestibular loss on the perceived sense of direction as commonly assessed with questionnaires. Moreover, disrupted cortical processing of multisensory visual and self-motion information associated with functional vestibular disorders may impair SOD abilities even when peripheral vestibular function remains intact.

Additionally, psychological symptoms such as anxiety and depression are often concomitant in patients with vestibular disorder and may act to worsen disorientation symptoms arising from vestibular dysfunction and prolong effective recovery [8, 44]. Moreover, 10% of people who report vestibular symptoms receive a primary diagnosis of an anxiety disorder and thus such disorders represent the most prevalent psychiatric association with vestibular-like symptoms. Accordingly, it has previously been shown that comorbid psychiatric symptoms act to worsen clinical outcomes in vestibular disorders and have consequently been shown to be the strongest predictors of persistent dizziness symptoms in some patients, as opposed to actual vestibular deficits [7, 22]. However, although general anxiety may be heightened in patients, common symptom questionnaires do not separate a spatial component of anxiety, which is likely more relevant to patients suffering from dizziness symptoms. Similarly, susceptibility to migraine also impedes recovery and aggravates vertigo in vestibular disorders [3], yet the effect of migraine symptoms on sense of direction remains unknown. In this study we implemented the Spatial Anxiety Questionnaire [34] to assess anxiety experienced by patients in spatially challenging environments, as well as other validated symptom questionnaires to establish the effect of psychological and migraine symptoms on orientational abilities in varying vestibular diagnoses.

Hence, our primary aim was to assess whether vestibular dysfunction impairs SOD ability by comparing orientation measure scores between a healthy control cohort and patients on the group level. Following aims were to then assess differences in SOD ability between varying vestibular diagnoses, and to explore the interactions of symptomatic dizziness, psychological comorbidity and migraine factors with SOD ability in patient and control groups.

2 Methods

3 Results

3.3 Performance on orientational measures is worse in patients compared to controls and is also linked with greater dizziness disability

A comparison of mean SBSOD score between healthy controls (N = 69) and all vestibular groups (N = 87) showed that mean score for patients was significantly lower on this scale (U = 2101.0, p < 0.001) (Fig. 1A). Similarly, patients scored significantly lower on the Object Perspective test (U = 630.0, p < 0.001) (Fig. 1C) and higher on the Spatial Anxiety Questionnaire (U = 1734.0, p < 0.001) compared to control subjects (Fig. 1B). SBSOD was significantly associated with spatial anxiety in vestibular patients (R² = 0.44, p < 0.001). Moreover, 68.8% of patients said they experienced no change in their SOD following their diagnosis, whilst 31.3% claimed it had worsened. Interestingly, patients who rated their SOD as worse scored significantly higher on the DHI (U = 507.5, p < 0.05) and Spatial Anxiety Questionnaire (U = 324.0, p < 0.001) and significantly lower on the Object Perspective test (U = 147.5, p < 0.05) compared to those who experienced no change in SOD ability (Fig. 2). No effect of age was observed on SBSOD or Object Perspective test score in either control subjects or patients (p > 0.05).

OPEN IN VIEWER

Fig. 1

Mean and 95% confidence interval for A) Santa Barbara Sense of Direction, B) Spatial Anxiety Questionnaire and C) Object Perspective test score in control and patient groups. Means compared via Mann-Whitney U test. ^*p < 0.05, ^**p < 0.01, ^***p < 0.001. N = 87 patients and N = 69 healthy controls except as indicated in C.

OPEN IN VIEWER

Fig. 2

Mean and 95% confidence interval on A) the Dizziness Handicap Inventory, B) Spatial Anxiety Questionnaire and C) Object Perspective test scales for patients who rated their SOD as having either worsened (N = 23) or remained unchanged (N = 63) since their diagnosis. Means compared via Mann-Whitney U test. ^*p < 0.05, ^**p < 0.01, ^***p < 0.001.

3.4 Deficits in orientational measure scores between vestibular patients and controls are limited to central/functional diagnoses

To assess whether SOD ability varied depending on aetiology of vestibular dysfunction, mean scores for main SBSOD, Spatial Anxiety Questionnaire and Object Perspective test variables were compared across patient groups and controls (Fig. 3). Most patient groups displayed a lowered SBSOD score in comparison to control subjects (Fig. 3A), however this effect was only statistically significant in patients with vestibular migraine (p < 0.05). For the Spatial Anxiety Questionnaire (Fig. 3B), PPPD and vestibular migraine groups alone scored significantly higher than controls (PPPD: p < 0.01, VM: p < 0.001). For the Object Perspective test (Fig. 3C), only patients with PPPD scored significantly worse than control subjects (p < 0.05).

OPEN IN VIEWER

Fig. 3

Mean and 95% confidence interval for A) Santa Barbara Sense of Direction scale, B) Spatial Anxiety Questionnaire and C) Object Perspective test score in vestibular groups and control subjects. Means compared via Kruskal Wallis test with post-hoc Dunn-Bonferroni correction. ^*p < 0.05, ^**p < 0.01, ^***p < 0.001.

The initial aims of this study were to evaluate the effect of vestibular dysfunction on sense of direction ability as quantified by the SBSOD and Object Perspective test measures, as well as secondary effects of dizziness, migraine and psychological symptoms across vestibular groups and control subjects. Overall, these results indicate that sense of direction and the related variables depicted in Fig. 1, are not so much affected by the loss of peripheral vestibular function, but by central or functional disorders (VM and PPPD). Moreover, perceived sense of direction shows association with experience of task-specific spatial anxiety, rather than generalised anxiety symptoms in both patient and control groups.

In terms of understanding the peripheral vestibular system’s contribution to sense of direction ability, we focus on patients with bilateral vestibular failure. Our diagnostic criteria included a 90% or greater reduction in vestibular function, as such this group could help in assessing whether normal vestibular function is essential for human sense of direction and orientation abilities. Perhaps unexpectedly, however, as Fig. 3 indicates, the BVF group do not show statistically significant differences on orientation measures with respect to control subjects, despite this group showing relatively low mean scores for both the SBSOD and Object Perspective test scales. Notably, considerable variation was observed in this group with individual-level inspection highlighting that the worst performing patient was an outlier 70 year old woman with multiple comorbidities, including diabetes, essential tremor with pendular pseudo-nystagmus [9], polypharmacy, antidepressant drugs and recurrent vertigo episodes fulfilling diagnostic criteria for vestibular migraine. Thus in general, our results indicate that bilateral vestibular hypofunction per se, does not appear to significantly impact sense of direction as assessed by the orientation measures used here.

Despite the small numbers of studies which have explored SOD and vestibular dysfunction, this finding is nonetheless reflected by previous work [13], in which patients with chronic bilateral vestibular failure perform similarly to control subjects in implicit and explicit perspective-taking tasks. Also supportive are results of experiments with purely vestibular navigation tests in normal volunteers, where performance levels in simple angular navigation tasks are not associated with SBSOD scores [49]. Conversely, a recent study reported an association between lower SBSOD score and saccular vestibular loss (absent C-VEMPs) [20]. We believe this finding, dictated by 3 subjects (out of 52) with bilaterally absent C-VEMPs, is limited by low subject numbers and the implausibility that short latency saccular reflexes are involved in SOD processing in humans (Nb recall that the sacculus senses linear acceleration along the longitudinal body axis, a kind of motion not relevant to human sense of direction). Since initial submission of our work, another study has reported deficits in Object Perspective test performance in a group of patients combining bilateral and unilateral peripheral vestibular lesions [18]. However, this work selected patients more on the basis of vestibular test results rather than clinical diagnosis, hence a comparison with our data is difficult. For instance, although the authors did not investigate PPPD patients, anxiety levels in peripheral patients were noticeably high, with two thirds classifying as moderately or severely anxious. Thus, this raises the possibility that many of these patients could suffer from PPPD in addition to the underlying vestibular disorder, with the former rather than the latter being associated with a deterioration in Object Perspective test performance.

Nevertheless, the non-significant finding in patients with BVF reported in the present study is further reinforced by the comparable performance of other peripheral vestibular groups, in particular those with Meniere’s disease, who also do not differ significantly from control subjects on main orientation measures. Thus, the lack of difference in SOD among peripheral disorders compared to control subjects likely reflect that these patients rely on visual and/or cognitive components for their sense of direction. Importantly, these components may remain intact or possibly become enhanced following vestibular loss and are likely more relevant to general SOD ability compared to vestibular motion information. This latter point is accentuated by the fact that we report non-significant findings on orientation tasks which are desk-based, which despite their correlation with real-world spatial tasks [25], are nonetheless completed in static environments with minimal vestibular contribution. Results from other studies investigating the effects of broader peripheral vestibular dysfunction on sense of direction specifically are lacking, but studies investigating cognitive spatial orientation abilities as assessed by mental rotation tasks report mixed results [10, 36]. However, although important to overall orientation ability, object manipulation and perspective-taking have been clearly separated as differing cognitive abilities [32]. As such, despite mixed evidence in the available literature, our results align with previous findings indicating that peripheral vestibular information likely contributes minimally to everyday sense of direction.

Consequently, the significant results observed in functional vestibular groups can likely be attributed to the complex neurobiology, prominent psychological factors, and the absence of central vestibular compensation commonly associated with these disorders [1]. The neurobiological mechanisms underlying PPPD remain unclear, however, numerous fMRI studies have consistently revealed functional abnormalities in the parieto-insular vestibular cortex (PIVC), which serves as an integration site for self-motion and visual information related to spatial orientation abilities [27, 42]. In vestibular migraine, cortical depolarisation originating from the occipital pole is thought to trigger trigeminovascular hypersensitivity responsible for headache symptoms, as well as abnormal activity of the central vestibular system [5]. Moreover, research has also revealed a reduction in grey matter volume within the PIVC in patients with VM, which has further been shown to have a negative association with DHI score [50]. Collectively, previous results suggest that functional abnormalities affecting the central vestibular system may disrupt the integration of multisensory inputs, precipitating ongoing symptoms of dizziness and impairing orientation abilities in VM and PPPD. Accordingly, performance on a vestibular-based orientation task has been found to correlate with the amount of long-term vestibular symptoms, including derealisation and depersonalisation, which are often reported by patients with VM and PPPD [30].

Patients with PPPD and VM in the present study showed comparable demographic characteristics to those reported in previous studies, being majority female with a mean age in the late 40s to 50s [16, 37]. Additionally, PPPD and VM groups alone showed elevated psychological symptom severity compared to healthy individuals in our study, also in alignment with prior cross-sectional studies which report highest prevalence of psychiatric comorbidity in patients with VM (49–65%), with much lower prevalence in peripheral disorders of BVF (24%) and BPPV (15%) [14, 17]. Heightened trait anxiety has been associated with disruption of spatial working memory in healthy individuals [35, 47], and has also been associated with increased sensitivity of the vestibular system [19]. Moreover, prior work has suggested that clinical outcome following acute vestibular neuritis is largely influenced by psychological factors in combination with increased visual dependence, not by the extent of peripheral vestibular recovery [11, 38]. While it has been hypothesised that a causal role of psychological factors in functional dizziness may relate to the projection of vestibular afferents to the parabrachial nucleus, a known regulator of emotional processing [4, 29], studies investigating the incidence of premorbid psychiatric disorder in functional dizziness disorders report mixed results [46]. In comparison to aforementioned studies, a critical finding here is that spatial-specific, as opposed to generalised anxiety symptoms, are more associated with SOD ability, which was consistent across the study population.

Correspondingly, correlations between spatial anxiety and both Object Perspective test and SBSOD scales have been shown in previous studies investigating healthy populations [24]. Similarly, yet distinct from sense of direction ability, a recent study demonstrated significant correlation between mental rotation ability and spatial, rather than trait anxiety symptoms and further indicated that the degree of spatial anxiety experienced is not inherently linked to individual trait anxiety [2]. Thus, this appears to support the existence of domain-specific spatial anxiety, which cannot be solely attributed to generalised anxiety symptoms.

In relation to vestibular dysfunction therefore, spatial anxiety may disrupt cognitive spatial abilities such as perspective-taking, as well as hyper-sensitising patients with functional disorders to vestibular inputs. This may lead to poor integration of multimodal sensory information and diminish subjective orientational ability. This effect is further supported by results from factor analysis conducted with the entire patient population as a whole, in which the primary component extracted loaded dizziness and psychological factors heavily, whereas in the secondary component, spatial anxiety showed clear separation from general symptomatic factors but showed strong inverse association with the SBSOD and Object Perspective test. This effect is essentially a mirror image of results in control subjects, as although in both groups the symptomatic and orientation factors remain segregated, the former is dominant in patients and the latter is dominant in control subjects. This is perhaps expected given the much stronger presence of symptoms in patients than in controls. Nevertheless, these results support the notion that experience of a specifically spatial component to anxiety is highly associated with subjective sense of direction in both patients with vestibular disorder and the general population. However, a possible limitation to this finding could relate to a degree of similarity between items on the SBSOD and Spatial Anxiety Questionnaire scales. Whereas the SBSOD relates to one’s perceived sense of direction ability, the Spatial Anxiety Questionnaire focuses on one’s experience of anxiety in various spatially challenging environments. Therefore, the similar constructs of these two measures may raise the possibility of shared variance between the reported scores.

5 Conclusion

In summary, this study aimed to investigate the impact of varying vestibular disorders as well as symptomatic dizziness and psychological factors on perceived sense of direction ability. Our results indicate that it is central and functional disorders (vestibular migraine and PPPD) rather than diagnoses with peripheral vestibular aetiology that account for impaired orientational abilities in patients when compared to control subjects. Notably, patients with bilateral vestibular failure do not show abnormal perceived sense of direction, indicating that this ability (as measured with the SBSOD scale) does not critically depend on normal levels of vestibular function. Moreover, spatial anxiety and orientational measures exhibit strong negative association with each other and show clear separation from symptomatic factors in both patient and control groups, suggesting that it is specifically spatial anxiety fears rather than generalised trait anxiety that is associated with poorer sense of direction ability. However, the relative contributions of the orientational and symptomatic components to overall sample variance show contrasting influences between groups, with the primary component comprising the dizziness and psychological symptom measures in patients. Although orientational scales used in this study have been shown to be reflective of real-world ability, future work should nevertheless aim to reaffirm these findings adding physical assessments of orientational ability in vestibular patients.