The value of vestibular graviceptive pathway evaluation in the diagnosis of unilateral peripheral vestibular dysfunction

Abstract Background Evaluation of vestibular graviceptive pathway (VGP) in patients with unilateral peripheral vestibular dysfunction (UPVD) has received increasing attention from researchers. The study aimed to investigate the value of VGP evaluation in the diagnosis of UPVD. Methods Ninety‐five UPVD patients were divided into attack and remission phase groups. VGP evaluation‐related indicators, including subjective visual vertical (SVV), subjective visual horizontal (SVH), head tilt, ocular torsion (OT), and skew deviation (SD), were measured, and their correlations with cochleovestibular function test results were analyzed. The possible etiologies of contralesional VGP (c‐VGP) were analyzed. Results Positive rates of SVV, SVH, OT, and SD were significantly higher, and the degrees of SVV, SVH, and OT were significantly greater in the attack phase group than the remission phase group. The sides with abnormal VGP evaluation results were correlated with the sides with hearing loss, abnormal caloric, and video head impulse test (vHIT) results. A total of 14 patients showed c‐VGP, and possible etiologies included contralateral benign paroxysmal positional vertigo (n = 4), bilateral hearing loss (n = 8), bilateral vHIT gain reduction (n = 1), autoimmune diseases (n = 6), vascular risk factors (n = 6), lacunar infarction (n = 3), and endolymphatic hydrops (n = 3). Conclusions Alterations in SVV, SVH, OT, and SD were noted in UPVD patients in different phases, which are presumed to be related to dynamic vestibular compensation; correlations between VGP evaluation results and cochleovestibular function test results indicate that VGP evaluation may be helpful for the diagnosis of the side affected in UPVD; the presence of c‐VGP may be related to bilateral labyrinth lesions or endolymphatic hydrops on the affected side; and the involvement of autoimmune mechanisms also deserves attention.


INTRODUCTION
Vertigo is a disorder of body's sense of balance and orientation in space, and an illusion of movement of the person or of the external world. Clinically, the etiologies of vertigo/vestibular disorders are complex, which involves multidisciplinary knowledge. In a population-based questionnaire study, approximately 20% to 30% of the population experienced dizziness/vertigo symptoms. A national survey of Germans showed that the lifetime prevalence of vertigo was 7.4% in adults aged 18-79 years, which was more common in women, with a female to male prevalence ratio of 2.7:1, and the prevalence of vertigo increased obviously with age (Strupp et al., 2020). In the United States, there are nearly 10 million visits to emergency departments each year for complaints of dizziness or vertigo, accounting for about 25% of all emergency department visits (Saber Tehrani et al., 2013).
In fact, damage to the unilateral peripheral vestibular pathway (including vestibular hair cells and vestibular nerve) caused by any physical or chemical factors can lead to asymmetric damage to the bilateral vestibular system, resulting in corresponding symptoms of imbalance, such as dizziness/vertigo, spontaneous nystagmus, severe nausea, and vomiting, this is called unilateral peripheral vestibular dysfunction (UPVD). A German multicenter study of 34,860 patients with dizziness/vertigo showed a high incidence of UPVD (approximately 9.1%) (Strupp et al., 2020). Patients with UPVD generally experience a rapid onset, most of these patients visit the hospital due to persistent dizziness/vertigo, and UPVD is often accompanied by clinical symptoms and signs such as spontaneous nystagmus, instability, nausea, and vomiting. For years, clinicians have relied mainly on semicircular canal function tests, such as caloric test, rotatory chair tests, and video head impulse test (vHIT), to determine the affected side of UPVD. A study has shown that up to 40% of patients with dizziness/vertigo showed unilateral canal paresis (CP) during caloric test (Goebel & Paige, 1989).
In recent years, with the rapid development of the anatomic basis of the vestibular system, clinical theories, and related research tools, the evaluation of vestibular graviceptive pathway (VGP) based on the impairment of utricular function and pathways in patients with UPVD has received more and more attention from researchers.
In clinical practice, VGP evaluations mainly include subjective visual vertical (SVV), subjective visual horizontal (SVH), head tilt (HT), ocular torsion (OT), and skew deviation (SD). Studies have found that 80%-94% of patients with UPVD showed SVV/SVH tilts toward the affected side, and about 10% of patients showed SVV/SVH tilts toward the unaffected side (Faralli et al., 2014). Hirvonen et al. (2011) andStrupp (2008) showed that 60% of patients with acute vestibular neuritis (VN) had HT to the affected side, and the inaccurate perception of gravity may be further exacerbated by head movement on the affected side. Faralli et al. (2021) showed that in most of the patients with superior VN, the SVV tilt values returned to normal range within 3-6 months, whereas OT remained abnormal after 1 year or even longer, suggesting that OT may be a good indicator of the entity of the residual peripheral otolithic lesion. In fact, it is still controversial whether OT or SVV recovers faster (Müller et al., 2016). Gufoni et al. (2020) found that the measurement of SD can determine when mechanical damage to the utricule may have occurred in patients with canalolithiasis.
In clinical practice, VGP evaluation is of great clinical value in the diagnosis of the affected side in patients with UPVD, but its clinical manifestations and value in the attack and remission phases of UPVD and the etiology of contraversive SVV/SVH deserve further investigations. Given this background, the present study investigated the changes in five VGP evaluation indicators (SVV, SVH, HT, SD, and OT) in 95 patients with UPVD (including 46 patients in the attack phase, and 49 patients in the remission phase) analyzed their correlations with related vestibular symptoms and vestibular function test results, so as to provide clinical evidence for the use of VGP evaluation in the diagnosis of UPVD.

vHIT
vHIT (Interacoustics, Middelfart, Denmark) was used to evaluate the function of the three pairs of semicircular canals. The instrument comprises an inertial measurement unit to measure head movements and an infrared camera to record eye movements. In a brightly lit room, patients were in the sitting position and wore an eye mask. There were instructed to fix their eyes on a target at eye level and at a dis-

VGP evaluation
2.4.1 SVV and SVH SVV was tested using an SVV measuring instrument (ZT-SVV-I, Shanghai ZEHNIT Medical Technology Co., Ltd., Shanghai, China). Subjects were seated upright with their heads in an erect neutral position and wore a contour mask that restricted the visual field to eliminate visual references. A yellow luminous bar with a length of 60 cm was projected on a black background of a visual field in front of patients (distance: 2 m), and an initial position of the bar was set at ±25 • . SVV and SVH were defined as 0 • based on the gravitational vertical and horizontal lines, respectively, and the two lines were used as vertical and horizontal coordinates. The circular field of view was divided into four quadrants, the tilts of the vertical and horizontal axes to the right/upward in quadrants 1 and 2 were designated recorded as positive, and tilt to the left/downward was designated as negative. Patients were asked to adjust the angle of the bar with a control joystick with an accuracy of ±0.1 • . Before the tests were started, patients were allowed to practice twice to familiarize themselves with the tests. In order to eliminate the influence of visual memory, the values were recorded from the third time of test. SVV and SVH tests were repeated seven times, respectively, and the average value was recorded.

HT
The patient was instructed to keep the body and head straight as far as possible before the examination, whereas the tester observed the posture of the patient. If the height of the patient's shoulders was not consistent, the patient was verbally prompted to adjust the height of the shoulders to keep the shoulders at the same level. After the adjustment, the tester measured the angle between the sagittal axis of the patient's head and gravity with the iPhone protractor, which was the HT degree (Brandt & Strupp, 2005). HT was measured separately by two experienced neurologists, and the average of the two measurements was taken as the final degree. Abnormal value was defined when the HT >2 • (Hirvonen et al., 2011).

OT
Fundus photographs were taken for both eyes using a nonmydriatic fundus camera. Before photography, patients were instructed to undergo 5 min of dark adaptation in a dark room to allow proper pupil dilation. Moreover, they were asked to keep their head in an upright position and look at a fixation target during photography. Patients were then asked to rest for 3 min with their eyes closed. After pupils returned to their normal size, the head position was kept unchanged, and photography for another eye was performed in the same way.
Angle between a line connecting the optic disc center to fovea center and a horizontal line passing through the optic disc center was measured using image analysis software. Two photographs were captured by each eye, and the final degree is the average of the two photographs. Abnormal OT was considered if the difference in torsional angle between the two eyes was ≥8.8 • (Choi et al., 2007).

SD
Maddox rod was placed in the front of the patients' right eye, and they were instructed to look at a light source at a distance of 33 cm with the both eyes. For the absence of SD, the dot and line that the patients saw were overlapped; for the presence of SD, the dot and line were separated from each other. After a prism was placed over the eyes, the patient complained that the separated dot and line are overlapped. The degree of prism was degree of SD (Green & Gold, 2021). Two experienced neurologists measured the SD separately, and the average of the two measurements was taken as the final degree. The time required for all 5 VGP evaluation is 20-30 min.

Ocular vestibular evoked myogenic potential (oVEMP)
oVEMP test was performed using the Interacoustics Eclipse system (Interacoustics, Middelfart, Denmark) in a conventional sound isolation room. Acoustic stimuli (intensity of 1000 dB nHL, short tone bursts of 500 and 1000 Hz) were delivered through insert earphones. The repetition rate was set to 5 times/sec with 200 stimuli repetitions/sweeps, and the recording window was 50 ms. The responses were bandpass filtered between 10 and 1000 Hz. The rise/fall time was 1 ms, and plateau time was 2 ms. During the test, patients were placed in a sitting position, and the local skin was cleaned. The recording electrodes were placed at about 1 cm below the midpoint of the bilateral infraorbital rim, the reference electrodes were placed 2-3 cm below the recording electrode, and the ground electrode was placed on the middle of forehead. Electrode impedance was maintained below 5 kΩ. Patients were instructed to maintain constant head position, elevate gaze to a fixed target about 1 m above the midline of the visual field at 30 • . During recording, blinking and eye closure should be avoided to maintain stable extraocular muscle tone.
Normal bilateral amplitude asymmetry ratio was defined as ≤0.33.
The normal oVEMP latency value defined by our laboratory was N13P18. No responses, amplitude asymmetry ratio falling outside of the normal range, or latency prolongation were considered abnormal results.

MRI protocol
A 3.0-T MRI imaging scanner (Magnetom Avanto, Siemens, Germany) with an 8-channel head coil was used to scan the entire temporal bone.
MRI protocol was as follows: T1-weighted imaging (T1WI), fast-spinecho (FSE) T2-weighted imaging (T2WI), and T2-DRIVE-HR plain in the transverse plane; enhanced T1WI in the transverse and coronal planes;  (Bernaerts et al., 2019). That is, when the saccule of the patient is equal or larger than the utricle, the patient is positive for endolymphatic hydrops.

Statistical analysis
Continuous variables were expressed as mean and standard deviation (mean ± SD), and comparisons of normally distributed and nonnormally distributed data were performed using independent sample

Clinical baseline characteristics of all patients included in the study
This study enrolled 95 patients with UPVD, including 46 patients in the attack phase of UPVD and 49 patients in the remission phase. In the attack phase group, there were 21 males and 25 females, with an average age of 52.22 ± 13.59 years. In the remission phase group, there were 19 males and 30 females, with an average age of 46.84 ± 13.19 years. There was no statistical significance in sex, age, and the affected side of UPVD between the two groups (Table 1).
Comparison of the positive rates of the five vestibular graviceptive pathway (VGP) evaluation indicators in patients with unilateral peripheral vestibular dysfunction (UPVD) between the attack and remission phase groups. (b) The association between the occurrence of ocular torsion (OT) and relapsing remitting phase of UPVD. HT, head tilt; nRR, non-relapsing remitting; SD, skew deviation; SVH, subjective visual horizontal; SVV, subjective visual vertical.

Results of qualitative analysis of VGP-related indicators
The positive rates of SVV, SVH, OT, and SD were significantly higher in the attack phase group than in the remission group (SVV: 65.2% vs.
2.0%, p = .012, Fisher's exact test). The positive rate of HT was higher in the remission phase group compared to the attack phase group, but the differences were not significant (26.1% vs. 38.8%, χ 2 = 1.738, p = .187, Pearson chi-square test, Figure 2a). In order to analyze the reasons for this phenomenon, we further divided the patients into relapsing Mann-Whitney test, Figure 3).

Correlations between VGP-related indicators
Correlations among all the five indicators in VGP evaluation were found in patients with UPVD in the attack phase, whereas correlations between SVV and SVH were found in patients with UPVD in the remission phase (r = .505, p < .001, Table 2).

oVEMP test results
There was no significant correlation between the sides with abnormal VGP evaluation results and abnormal oVEMP results in either the attack phase group (χ 2 = 2.741, p = .626, Fisher's exact test) or the remission phase group (χ 2 = 2.837, p = .611, Fisher's exact test).

Clinical characteristics of UPVD patients with contralesional VGP (c-VGP)
From the abovementioned correlation analysis between the VGP results and the caloric test results, we found that 14 (14/95, 14.7%) patients showed contralesional VGP (c-VGP), that is, the sides with abnormal VGP evaluation results were opposite to the sides with abnormal caloric test results. In order to analyze the reasons for the presence of c-VGP, we divided patients into a c-VGP group and an ipsilesional VGP (i-VGP) group and conducted further multidimensional evaluation.

3.4.1
Clinical characteristics of UPVD patients with c-VGP Among the 14 UPVD patients with c-VGP, 12 (85.7%) patients were in the attack phase, and 2 (14.3%) were in the remission phase. There was no statistical significance in the degrees and positive rates of VGP evaluation indicators, the degree of spontaneous nystagmus, and CP value in UPVD patients in both the attack and remission phases between the c-and i-VGP groups (

DISCUSSION
Previous studies (Brandt & Dieterich, 1995;Choi et al., 2007;Korda et al., 2022;Strupp, 2008;Zhao et al., 2017) have shown that for patients with UPVD in the attack phase, the positive rates of SVV, SVH, HT, OT, and SD were 50.6%−94%, 54.1%−91%, 4%−37%, 19%−82%, and 14%−24%, respectively, whereas for patients with UPVD in the remission phase, the positive rates of SVV, SVH, HT, OT, and SD were 0%−25%, 0%−22%, 0%−20%, 0%−20%, and 0%−5%, respectively, and these findings suggest that SVV, SVH, OT, and SD tend to return to normal range in patients with UPVD in the remission phase. In the present study, the results from VGP evaluation are quite similar to the abovementioned findings, and our results showed high positive rates of SVV and SVH. A previous study (Conrad et al., 2014) showed that in contrast to HT, OT, and SD, the body requires multiple peripheral, central vestibular pathways, and multisensory integration at high-level nerve center to maintain the accuracy of SVV/SVH, such as vertical semicircular canal pathways, utricle-mediated pathways, medial longitudinal fascicle pathway, brachium conjunctivum pathway, and ipsilateral vestibulothalamic tract pathway, as well as the integration of vestibular, proprioceptive, and visual information. We hypothesize that SVV/SVH is more sensitive and vulnerable than other VGP indicators, that is, damage at any site can lead to SVV/SVH abnormalities.

F I G U R E 5 Correlations between vestibular graviceptive pathway (VGP) evaluation results and semicircular canal function test results: (a)
There were significant correlations between the sides with abnormal caloric test results and the sides with abnormal VGP evaluation results; (b) there were significant correlations between the sides with abnormal VGP evaluation results and the sides with abnormal video head impulse test (vHIT) results.
In this study, dynamic changes in SVV, SVH, OT, and SD were found in UPVD patients in both the attack and remission phases, whereas obvious change in HT was not noted. We speculate that this may be due to the following two reasons: (1) The body may be more sensitive to ocular position compensation and cortical perception than to vestibulo-collic reflex (VCR) compensation. A study of patients with unilateral vestibular deafferentation after vestibular schwannoma resection conducted by Mantokoudis et al. (2014) found that compensation for SD was more quickly compared to dynamic VOR compensation. Aleisa et al. (2007) showed that compared with dynamic VOR, mice showed less compensation for VCR after unilateral labyrinthectomy. We specu-   head to tilt to the affected side (Kent et al., 2010). In the present study, we found that the relapsing remitting phases of UPVD were associated with HT. Asama et al. (2012) showed that vestibular compensation for VCR occurred in patients with chronic dizziness, further increasing proprioceptive afference from the neck muscles to the vestibular nucleus, they believe that the asymmetry of the neck muscle tension could reflect the bilateral asymmetry of vestibular function, and the increase in the asymmetry of the neck muscle tension is intended to compensate for postural stability, that is, vestibular compensation can cause HT in patients after the peripheral vestibular function is impaired on one side. Further dynamic studies with a large sample size are needed to explore the relationship between HT and disease duration.
In terms of correlation, 5 VGP evaluation indicators showed a pairwise correlation in the attack phase group, whereas in the remission phase group, only SVV and SVH were correlated, which reflected the dynamic changes of the VGP evaluation indexes from the side, that is, the VGP evaluation indicators that respectively reflect vestibulecervical, vestibule-ocular, and cortical perception have different compensation speeds in different periods. In previous studies, it is generally believed that SD and HT recover quickly and usually improve within 4-8 weeks (Hirvonen et al., 2011;Strupp, 2008). There is controversy over the recovery speed of SVV and OT. Studies have shown that both SVV and OT can be improved between 1 and 6 months (Choi et al., 2007). Some studies suggest that OT has a longer recovery time than SVV and can last for a year (Faralli et al., 2021). In addition, some studies also believe that SVV can exist in chronic UPVD in a small The patient complained of obvious tinnitus symptoms in the left ear since the onset of the disease (there was no significant difference in hearing loss in both ears). Therefore, combined with clinical symptoms, even if the PTA hearing loss in the left ear did not reach the WHO standard classification of 20 dB, we also considered the patient to be qualified for bilateral hearing loss.
TA B L E 5 Comparison of age, the prevalence of vascular risk factors and immune abnormalities in unilateral peripheral vestibular dysfunction (UPVD) patients between the contralesional VGP (c-VGP) and ipsilesional VGP (i-VGP) groups  Kim and Hong et al. (2008)  peroxidase antibody (TPOAb) and thyroglobulin antibody were found to be positive in 90% of patients with Hashimoto's thyroiditis (Gunes et al., 2017). Circulating autoimmune antibodies can cause damage to the labyrinth and connective tissues in various organs throughout the body. Chiarella et al. (2014) showed that up to 44.7% of patients with Hashimoto's thyroiditis showed unilateral CP during caloric test, and 55.2% of the patients showed abnormal VEMP. Gawron et al. (2004) showed that in patients with Hashimoto's thyroiditis, auditory nerve and brainstem neural conduction in the brain auditory evoked potentials were both affected, and TPOAb titers were positively correlated with the extent of central hearing disorder. It can be seen that autoimmune abnormalities associated with Hashimoto's thyroiditis might be one of the possible causes of c-VGP in UPVD patients.
The study has some limitations. This was a single-center study with small sample size, so there is a possibility of selection bias. In addition, dynamic follow-up study on UPVD patients was not performed in the study. Further studies with larger sample size and dynamic follow-up are needed to explore the clinical application value of VGP evaluation in the diagnosis of UPVD.

CONCLUSIONS
Our findings suggest that (1) alterations in SVV, SVH, OT, and SD were noted in UPVD patients in the acute attack and remission phases, which are presumed to be related to dynamic vestibular compensation; (2) there were significant correlations between VGP evaluation results and semicircular canal function test, audiological test results, indicating that VGP evaluation might be also useful to determine the affected side in UPVD patients; (3) the presence of c-VGP in UPVD patients may be related to bilateral labyrinthine lesions or endolymphatic hydrops on the affected side, and the involvement of autoimmune mechanisms also deserves attention.

AUTHOR CONTRIBUTIONS
Xiao-hong Ba, Xu Yang contributed to the conception and design of the study. Tong-tong Zhao collected the data, analyzed the results, and drafted the manuscript. Meng-lu Zhang, Yu-fei Feng, Qian-qian Wang, and Ning Song collected the clinical data and analyzed the results. All authors contributed to the article and approved the submitted version.

CONFLICT OF INTEREST STATEMENT
The authors declare no competing interests.

DATA AVAILABILITY STATEMENT
The data used in this study are available from the corresponding author, upon reasonable request.