Postural instability (PI) is one of the most distressing motor symptoms in PD and significantly increases the risk of falls (18, 19). Approximately 70% of patients with PD fall at least once annually (2, 20); moreover, this instability is known to increase with PD progression. PI occurs in approximately one-third of patients 2 years after the diagnosis of PD, rising to 71% after 10 years, and reaching 92% after 15 years (21). Vestibular signaling plays a crucial role in processing ego-motion information and regulating posture and balance (10), with vestibular dysfunction being an independent risk factor for falls in patients with PD and animal models (2–4). Six sensory integration tests using the NeuroCom Dynamic Postural Balance Instrument revealed that disrupted balance in patients with PD resulted from the inability to effectively interpret vestibular information independent of visual and proprioceptive integration and the loss of nigrostriatal dopamine during disease progression (3).

Altered vestibular-evoked myogenic potentials (VEMPs) increase as PD progresses to advanced stages and correlate with PI (22, 23), suggesting that impaired vestibular activity is a critical underlying factor. A prospective one-year follow-up study found that neurovestibular dysfunction might predict falls in at-risk patients with PD and imbalanced posture (12), suggesting that a baseline assessment using VEMPs may help predict future PI occurrence. Overall, PI in PD is associated with vestibular dysfunction.

Regarding anatomical localization, vestibular dysfunction in PD is associated with changes in the substantia nigra and pedunculopontine nucleus (PPN)–thalamic cholinergic innervation. For example, altered levels of vesicular acetylcholine transporters in the medial geniculate body are associated with PI and gait abnormalities in patients with PD (24). Cholinergic nerve endings in the vestibular brainstem nuclei include the caudal medial vestibular nucleus (25). Therefore, vestibular sensory information processing could be modulated by cholinergic signaling, further implicating vestibular dysfunction in PI development.

Pisa syndrome (PS), a disabling complication of PD, is defined as a specific set of recoverable postural changes involving ≥10° of lateral trunk flexion (LTF) and vestibular defects (13, 26, 27). The estimated prevalence of PS in patients with PD (PDPS) is 7.4–10.3% (28); these patients are more likely to have disordered balance and experience falls than those without PS (27). Therefore, PS severely reduces the QoL of patients and can lead to fractures and death, especially during advanced stages. As the primary system involved in regulating postural balance, vestibular dysfunction might be associated with PS development in patients with PD (29, 30).

Unilateral damage to the vestibulospinal reflex arc in experimental animals causes an imbalance in the descending motor regulation system of the spinal cord, resulting in scoliosis (31). Bilateral vestibulospinal reflexes are defective in patients with PDPS, who tend to have more severe cervical VEMP (cVEMP) abnormalities than healthy controls, suggesting an association between the vestibulospinal pathway and PS pathophysiology in PD (14). Scocco et al. (15) found that 14 of 21 patients with PD and LTF had pathological subjective visual vertical (SVV) perception; Gandor et al. (16) reported similar results. Therefore, vestibular balance disorders might be involved in LTF pathophysiology. Vitale et al. examined 11 patients with PD and LTF and found vestibular hypofunction in all of them (13). These findings suggested a potential connection between the postural changes in PD and a combination of vestibular dysfunction and altered somatosensory integration, with vestibular dysfunction driving the mechanisms leading to scoliosis in PD.

Freezing of gait (FoG) is a unique and disabling clinical phenomenon characterized by brief episodes of inability to step forward or by extremely short steps that typically occur on initiating gait or turning while walking. Symptoms of FoG in patients with PD are also associated with vestibular dysfunction. Approximately 84.6% of patients experiencing FoG failed the modified Romberg 4 subtest, which focuses more specifically on altered vestibular function (17). Thus, vestibular processing deficits might pathophysiologically correlate with FoG in patients with PD.

Postural sensory processing, particularly of vestibular information, is poorer in patients with FoG, possibly due to impaired central processing of vestibular signals (32). Primates have a high vestibular response in the PPN (33). In addition, PPN connectivity can be enhanced via galvanic vestibular stimulation (GVS) in patients with PD (34), while rotational stimulation of the vestibular system might alleviate FoG (35). A study of four patients with PD who had undergone bilateral PPN and subthalamic (STN) electrode implantation found that deep brain stimulation (DBS) of the PPN improved vestibular perceptual thresholds, further confirming the response of PPN neurons to vestibular stimuli (36). Thus, the PPN might be a vestibular signal processing center within the brainstem, and its stimulation might improve posture and gait in patients with PD by modulating vestibular signals, thereby reducing the risk of falling. However, definitive proof of a causal relationship between FoG and impaired vestibular processing is lacking and further investigation is required.

The vestibular system controls not only gait and posture but also eye movements, mainly through the vestibular-ocular reflex (VOR), which detects head movements and maintains image stability on the retinal fovea. Pathological changes involving the formation of Lewy bodies induced by the immune response to α-synuclein, the vestibular nucleus, oculomotor nucleus, and basal and upper ganglia affect saccadic and microsaccadic eye movements (37). Coincidentally, evidence of VOR abnormalities have been reported in many early studies of PD (4, 38).

A study of eye movements in 35 patients with idiopathic PD using videonystagmography (VNG) revealed visual and vestibular motoneuron abnormalities (39). Similarly, patients with PD exhibited poorer stereopsis and impaired oculomotor behaviors, as assessed using a three-dimensional active shutter system and Tobii Eye Tracker, respectively (40). Large meta-analyses have also confirmed the occurrence of increased response latencies in pro- and anti-saccade tasks in patients with PD (41, 42). A cross-sectional study of oculomotor performance found prolonged saccadic latencies, poorer saccadic accuracy, and lower gain in smooth pursuit eye movement (SPEM) in patients with de novo PD compared with those in healthy individuals (43). In addition, vestibular rehabilitation (VR) was shown to help improve the VOR gain (44). Therefore, eye movement abnormalities in patients with PD appear to be closely associated with vestibular system changes. However, extensive case–control studies are needed to confirm this.

The prevalence of dizziness complaints among patients with PD (48–68%) is twice as high as that among older individuals (20–30%) (1, 45). A case–control study reported that patients with de novo PD sometimes experience dizziness for many years before disease diagnosis (46). Dizziness in patients with PD is most often attributed to orthostatic hypotension (OH) (47, 48). Accordingly, various cross-sectional studies have shown that 30–50% of patients with PD have OH (49, 50). However, many patients with PD without OH also complain of dizziness. This type of dizziness is called non-specific dizziness and refers to a symptom of feeling dizzy or lightheaded without a specific identifiable cause. Of note, non-specific dizziness is the second most prevalent type of dizziness after OH. A retrospective study including 80 patients with early-stage PD (disease duration ≤5 years) found that 37 (46.3%) presented with dizziness, which was non-specific in 11 (29.7%) of the cases (1).

Although the mechanism leading to non-specific dizziness in patients with PD is unclear, it may be affected by vestibular dysfunction. A prospective study of 84 patients with PD with and without OH found greater impairment of vestibular function in patients with dizziness compared with those without dizziness (51). An electronystagmographic assessment of 30 patients with PD found vestibular disorders in 83% of them, with a 43.3% prevalence of dizziness despite the absence of complaints (52).

In summary, dizziness is prevalent among patients with PD; the symptoms of non-specific dizziness might be closely associated with vestibular system dysfunction. Thus, targeted vestibular therapy or improving dysfunctional vestibular-related neural circuits might help relieve non-specific dizziness in patients with PD.

Vestibular dysfunction might also be linked with sleep disorders. A study of nine adults with abnormal sleep patterns and shorter sleep durations uncovered greater bilateral vestibular hypofunction in these patients than in healthy controls (53). A cross-sectional study of 20,950 patients also found that 30% of individuals with vestibular vertigo had abnormal sleep durations; those with vestibular symptoms were more likely to experience insomnia or lethargy (54). Changes in VEMPs were found to directly correlate with rapid eye movement (REM), sleep behavior disorder, and PI (22). cVEMP abnormalities directly correlated with sleep scores in patients assessed using the REM Sleep Behavior Disorder Screening Questionnaire (55). Additionally, a small randomized controlled trial of 20 adults found that repeated electrical vestibular stimulation administered for 30 min/day improved total Insomnia Severity Index scores (56). These results indicated that vestibular pathways could project into multiple sleep and circadian-regulating nuclei in the brainstem and hypothalamus.

Mechanistically, the vestibular system regulates circadian rhythms and influences sleep behaviors by converging inputs from the visual and somatosensory systems (57). The higher centers of the vestibular system include many subcortical and cortical structures (58), which correlate with the function of the nerve center that regulates sleep; the hypothalamic circadian rhythm in animals is modulated through a neuroanatomical pathway between the medial vestibular nuclei and suprachiasmatic nucleus (59). In addition to sending projections to the cerebellar and brainstem nuclei, glutamatergic neurons in the vestibular nucleus project to the sleep–wake centers associated with brain regions that might be involved in sleep regulation (60).

Although the precise molecular mechanisms remain unknown, orexin expression increases after sleep deprivation (61). Interestingly, sleep regulation has been reported to depend on the activity of orexin-producing neurons in experimental animals. Orexin-producing neurons are closely associated with the vestibular system (62), supporting the correlation between the vestibular system and sleep regulation.

Mood disorders are a group of psychological disorders predominantly characterized by abnormal emotional regulation. Anxiety, fear, depression, and other neuropsychiatric symptoms are common in the general population and among patients with PD, although the underlying neurobiological mechanisms are complex and unclear (63). Many patients with impaired vestibular function also have emotional disorders, including those related to anxiety and fear (64). These relationships are not unidirectional, as mood disorders such as anxiety can lead to vestibular dysfunction and vice versa.

Ventricles synthesize adrenocorticotropic hormones, participate in processes associated with stress and anxiety, act on the vestibulolateral nucleus, and help regulate postural balance. Thus, stress, anxiety, and balance control are closely related to vestibular nuclei activity (65). Clinical findings have shown that caloric vestibular stimulation modulates mood and affective control (66); for instance, unrealistic optimism was selectively reduced during cold caloric stimulation of the left ear (67). The underlying mechanism might involve the vestibular cortex, and laterality may also be significant.

The right hemisphere is superior in dealing with negative emotions (68, 69). Similarly, vestibular cortical regulation is also affected by hemispheric lateralization, with the right hemisphere being more active than the vestibular cortex projection area in the left hemisphere in right-handed individuals (70). Hence, the right hemisphere might specialize in processes associated with interpreting or regulating negative emotions. More specifically, the right prefrontal region plays a role in negative emotional regulation associated with vestibular stimulation (71, 72). These findings indicated that cerebral hemisphere lateralization plays essential roles in vestibular and emotional processing, thus providing a theoretical basis for the interaction between the vestibular system and emotional processing centers.

Vestibular cognitive (also known as advanced vestibular) function is associated with visuospatial interaction, attention, executive function, and memory; accumulating evidence have supported a link between vestibular disorders and cognitive impairment in humans (73–75). Thus, cognitive decline in patients with PD might be associated with vestibular dysfunction.

Patients with PD exhibit visuospatial processing capacity deficits associated with aberrant neural circuitry in the frontal basal ganglia (76). However, this alone does not fully explain the cause of visuospatial deficits in PD. Similar to the basal ganglia, the vestibular system is also involved in spatial processing; vestibular dysfunction has been identified as the cause of decreased visuospatial competence in patients with PD (77). Patients with PD are more likely to experience cognitive dysfunction if they have gait and balance disorders; moreover, both symptoms worsen with disease progression (78). Dizziness might also be associated with cognitive decline in PD. For example, dizziness was closely associated with low Montreal Cognitive Assessment scores in patients with early-stage PD (1). Conversely, patients with refractory dizziness experienced significant improvement in cognitive function and dizziness-related indicators following VR therapy (79).

Vestibular function and cognitive processing might be linked to hippocampal activity (80). The hippocampus is involved in emotional processing closely associated with cognitive processes such as spatial memory (81). The human posterior hippocampus is involved in information processing and spatial memory, whereas the ventral hippocampus is responsible for emotional regulation (82).

The mechanisms underlying cognitive dysfunction in PD appear to be inextricably linked to the function of the vestibular system. However, further investigation is required to delineate these relationships.

Vestibular signal transmission and reception frequently coincide with those of other types of sensory information; central vestibular processing is promiscuous, contributing to its ubiquitous nature. Hearing loss has recently been recognized as an additional non-motor symptom of PD (83, 84). The factors affecting hearing loss are multifaceted, and common diseases such as otitis media with effusion also play a role in vestibular dysfunction (85). Hearing and balance function have certain neural connections and pathophysiologic mechanisms with motor control. Thus, neurodegeneration in PD may affect these common neural pathways, leading to hearing and balance problems. Although other non-motor symptoms, such as perception deficits, urinary problems, and cardiovascular and sexual dysfunction, are seemingly unrelated to vestibular function, some studies have suggested associations between these symptoms and VEMPs (7, 22). Thus, symptoms associated with vestibular system dysfunction are prevalent in patients with PD. While the underlying mechanisms are not fully understood, current pathologic, physiologic, and anatomic evidence support the idea that patients with PD have vestibular dysfunction (Table 2).