Abstract
Background and Purpose:
Abnormal balance is poorly responsive to dopaminergic therapy in Parkinson's disease (PD). Decreased vestibular efficacy may contribute to imbalance in PD. The purpose of this study was to investigate the relationship between vestibular measures of dynamic posturography and imbalance in PD while accounting for confounder variables.
Methods:
106 patients with PD underwent dynamic posturography for the 6 conditions of the sensory integration test (SOT) using the Neurocom Equitest device. All SOT measures, nigrostriatal dopaminergic denervation ((+)-[11C]DTBZ PET), brain acetylcholinesterase ([11C]PMP PET), age, duration of disease, cognitive and parkinsonian motor scores, and ankle vibration sensitivity were used as regressors in a stepwise logistic regression model comparing PD patients with versus without imbalance defined as Hoehn and Yahr (HY) stage 2.5 or higher.
Results:
The presence of imbalance was significantly associated with vestibular ratio COP RMS (P=0.002) independently from visual ratio COP velocity (P=0.012), thalamic acetylcholinesterase activity (P=0.0032), cognition (P=0.006), motor severity (P=0.0039), age (P=0.001), ankle vibration sensitivity (P=0.0008), and borderline findings for somatosensory ratio COP velocity (P=0.074) and visual ratio COP RMS (P=0.078). Nigrostriatal dopaminergic denervation did not achieve significance.
Conclusions:
The inability to efficaciously utilize vestibular information to retain upright stance is a determinant of imbalance in PD independent from visual and somatosensory processing changes and nigrostriatal dopaminergic losses. Thalamic, but not cortical, cholinergic denervation incrementally predicted balance abnormality. Further research is needed to investigate an intrinsic role of the cholinergic thalamus in multi-sensory, in particular vestibular, processing functions of postural control in PD.
Keywords: Acetylcholine, balance, dopamine, Parkinson's disease, PET, vestibular
1. Introduction
Vestibular system function is multi-faceted and can be defined as the integration of both peripheral vestibular labyrinth and central vestibular system functions. Vestibular dysfunction has been implied in the pathophysiology of parkinsonian imbalance for many years. Studies examining the specific contribution of vestibular sensory function to postural instability and gait disorder (PIGD) motor features in PD are, however, inconsistent (see Smith, 2018, for review [1]). For example, a study using electronystagmography and caloric testing found evidence of higher frequency of vestibular dysfunction in patients with PD compared to healthy controls [2]. Importantly, vestibular dysfunction was associated with impaired balance in the PD patients. More recent research by de Natale et al., found that vestibular-evoked myogenic potentials (VEMPs) were altered in patients with PD compared to controls, with the severity of these alterations progressively increasing with advancing disease [3]. These alterations in VEMPs were correlated with greater postural instability, suggesting that impaired vestibular activity may be a critical factor underlying abnormal balance in PD [4].
However, there are also conflicting reports; for example, findings by Pastor et al. showed that postural sway in more severely affected PD patients was increased in response to galvanic vestibular stimulation compared to mildly affected PD patients [5]. A major limitation of the literature in this field and a possible explanation for the divergent literature findings is that analyses in these studies were not adjusted for confounder effects of visual or somatosensory postural processing functions, cognition or the severity of striatal dopaminergic losses. Therefore, the major objective of this study was to examine the association between vestibular measures of dynamic posturography and abnormal balance in PD while controlling for confounder variables. We previously have shown that decreases in striatal dopamine associated with normal aging do not impair the ability of the vestibular control system indicating that this likely represents a non-dopaminergic function [6]. Therefore, we hypothesized that impaired vestibular efficacy contributes to balance dysfunctions in PD independent not only from somatosensory and visual postural control mechanisms, but also from the severity of nigrostriatal dopaminergic losses.
2. Subjects and Methods
2.1. Subjects
We enrolled 135 participants of both normal controls and Parkinson’s patients. The diagnosis of PD was confirmed by the UK Parkinson’s Disease Society Brain Bank clinical diagnostic criteria [7] and also consistent with a typical pattern of nigrostriatal dopaminergic denervation on [11C]dihydrotetrabenazine (DTBZ) vesicular monoamine type 2 (VMAT2) PET imaging in all subjects. Participants with a history of Meniere's disease or recent history of acute vertigo indicative of benign positional vertigo were not eligible for the study. Although most of the patients were on dopaminergic drugs, none were on anti-cholinergic or cholinergic drugs. Large fiber lower extremity vibratory sensory information was based on the degree of vibratory sensation detection at the lateral malleolus (right and left averaged) using the biothesiometer device (Bio-Medical Instrument company), which is an electric tuning fork providing superior accuracy compared to a manual tuning fork in terms of determining vibratory threshold. A neuropsychological test battery was also performed.
The medical ethics committee at the University of Michigan (IRB. number HUM00022832) approved this study and informed consent was obtained in all study participants. Subjects were recruited from a PET clinical correlation study in patients with PD (ClinicalTrials.gov Identifier: NCT01565473). Eligibility for this study was limited to study participants with dopaminergic PET evidence of nigrostriatal degeneration and without evidence of contra-indication for MRI or PET imaging and who were not taking anti-cholinergic or cholinesterase inhibitor drugs. We excluded participants with mass lesion on MRI or a history of current or recent vertigo. A total of 160 patients were recruited. Five patients were excluded because of a normal dopamine PET scan and 5 because of other screen failure or withdrawal. The current study presents data on all subjects who were able to complete the Neurocom SOT test (i.e., being able to stand independently) and not having a recent or current history of vertigo.
2.2. Dynamic posturography using the Sensory Organization Test
The study participants completed the Equitest posture test platform Sensory Organization Test (SOT) (Neurocom International, Inc. Clackamas, OR, USA) while on their usual dopaminergic medications. The dynamic posturography SOT consists of six sensory testing conditions to assess the integration of proprioceptive, visual, and vestibular functions during upright standing. The various conditions examine the individual’s ability to maintain standing using combinations of normal, altered (sway-referenced) or missing visual and proprioceptive cues during testing sessions lasting a half minute with short breaks in between. The SOT is considered the criterion standard test for assessment of sensory processing during postural control and has been validated in children and elderly populations, as well as in patients with vestibular disorders [14, 28].
The Equitest posture test platform is comprised of a dual force plate and a visual surround. Both the force plate and the visual surround can be sway-referenced. This means that the force plate can translate forward and backward along the y-axis parallel to the floor at a maximum linear velocity of 15.24 cm/sec and rotate about the x-axis at a maximum angular velocity of 50 °/sec to promote ankle dorsiflexion and plantarflexion. In a similar way, the visual surround can tilt about the x-axis at a maximum angular velocity of 15 °/sec [13]. The dual force plate of the Equitest postural platform consists of four force transducers to measure the vertical forces for each limb with a fifth transducer to measure shear between the two plates. Either a hip or ankle postural strategy may be used to maintain balance during the SOT, meaning a reactive torque may be generated about either the ankle or the hip joint in response to the degree of perturbation induced by the sway referenced motion of the force plate.
The center of pressure (COP) is defined as the location of the vertical ground reaction vector on the force platform and is a measure of the body’s ability to keep the center of gravity (COG) over the base of support. It consists of a bivariate distribution of points in the medial-lateral and anterior-posterior directions. Total COP excursion and COP RMS were calculated to estimate postural sway. Postural steadiness COP functions were defined as time domain measures of distance, area (RMS), velocity, and frequency-domain measures of spectral magnitude or distribution [12]. Therefore, the following three COP measures were selected as primary outcome variables: COP RMS, COP velocity and COP frequency. The COP RMS is the root-mean-square distance from the geometric mean of the COP. The COP speed measures the time rate of change of the COP excursion, which is the total distance traveled by the COP. The COP frequency measures the power spectral density of the standing sway [27]. The COP-based measures are derived for each of the six SOT conditions. Higher values of the COP measures indicate greater postural sway.
In SOT 1 and 2, the person stands quietly with eyes open (condition 1) and closed (condition 2). This establishes whether sway increases when visual cues are removed and determines how effectively the participant makes use of somatosensory input. In SOT 3, the person stands with their eyes open; the visual surround is sway-referenced and visual cues are no longer accurate. In SOT 4, the support surface is sway-referenced; thus, somatosensory cues are no longer accurate. SOT 5 is performed with eyes closed and a sway-referenced support surface. SOT 5 is the vestibularly challenged condition because it determines how the person uses vestibular cues when visual cues are removed, and somatosensory cues are no longer accurate. In SOT 6, the support surface and visual surround are simultaneously sway-referenced, which indicates if the person relies on visual cues even when they are no longer accurate. Integrated measures derived from the 6 SOT functions have been previously reported [11]. The previous study showed that the postural sway-related factor for SOT conditions #1-3 correlated with thalamic cholinergic innervation. SOT 5 scores were used for this study given the specific hypothesis of vestibular efficiency and balance functions in PD.
Sensory ratios were computed by dividing the COP measures for SOT2, SOT4 and SOT5 by those calculated for SOT1 [13]. In addition, a sensory ratio was calculated for visual preference (PREF) by dividing the sum of the COP measures for SOT3 and SOT6 by the sum of the COP measures for SOT2 and SOT5. These ratios provide an estimate of how well the patient can utilize a particular sensory system to maintain balance while other sensory cues are either altered absent [14]. For a more comprehensive explanation of SOT conditions and variables, please refer to Prieto et al. [12].
2.3. Imaging techniques
Our key imaging modalities are nigrostriatal dopaminergic assessment using [11C]-DTBZ PET and acetylcholinesterase [11C]-PMP PET. Loss of dopamine is the key and most consistent neurotransmitter change in PD compared to more variable and more limited losses of other monoamines and acetylcholine. The inclusion of [11C]-DTBZ PET is unique in a postural control study in patients with PD as it allows direct statistical adjustment for the biological severity of this disease based on nigrostriatal losses and hence will show dopaminergic system-independent effects of vestibular efficacy changes in PD. The inclusion of acetylcholinesterase PET is important given prior observations that thalamic cholinergic losses contribute to sensory processing during postural control independent from nigrostriatal dopaminergic losses [11].
A 3 Tesla Philips Achieva MRI camera (Philips, Best, The Netherlands) and an ECAT Exact HR+ PET system (Siemens Molecular Imaging, Inc., Knoxville, TN) were used as reported previously [15]. These two scans were performed on the same day in the vast majority of the participants. It was only when there was a radiochemical synthesis or other technical failure that a scan was performed within a few days of each other.
[11C]DTBZ and [11C]PMP were prepared as previously reported [16]. A 60-minute bolus/infusion protocol was used for [11C]DTBZ PET imaging (15 mCi) [17]. Dynamic PET scanning (70 minutes) using an i.v. bolus dose of 15 mCi was used for [11C]PMP imaging. [11C]DTBZ PET imaging was performed in the dopaminergic medication off state in the morning.
2.4. Image Analysis
All imaging frames were spatially coregistered within subjects using rigid-body transformation to decrease subject motion effects [18]. Interactive Data Language image analysis software (Research systems, Inc., Boulder, CO) was used to trace volumes of interest (VOI) on MRI images to include the striatum (putamen, caudate nucleus) and thalamus. A total neocortical VOI was defined using semi-automated threshold delineation of the MRI cortical gray matter signal [15]. [11C]DTBZ distribution volume ratio (DVR) of the bilaterally averaged striatum was determined based on the Logan plot graphical analysis method with the supratentorial cortex as reference region [19]. Thalamic and cortical acetylcholinesterase (AChE) [11C]PMP hydrolysis rates (k3) were determined based on the striatal volume input function [20].
2.5. Data Analysis
Abnormal balance was clinically defined as HY stages greater than or equal to 2.5 using the modified HY staging schema. A single examiner performed the retropulsion test throughout the study, making it a reliable measure of balance. We chose this definition because the HY is the most commonly used tool for defining disease severity in PD [26].
In order to assess whether PD participants with abnormal balance differ significantly from PD participants with normal balance or normal controls on any of the non-SOT covariates, a series of ANOVA models predicting the relevant covariates from group membership were conducted. Planned linear contrasts were used to compare both the PD group with normal balance and the control group against PD with abnormal balance. Model p-values were adjusted for multiple comparisons using Holm’s method [39]. Linear contrasts were only presented as significant if the associated model was significant, as defined by α = 0.05. Since levodopa equivalent dose (LED) and duration of PD are only applicable to the PD groups, normal controls were omitted from the ANOVA models associated with these variables, and only PD participants with and without abnormal balance were contrasted. To test for differences in sex across our three groups, a contingency table of sex by group was constructed, and analyzed using Fisher’s exact test. A mosaic plot was used to examine the distribution of sex across participant groups.
Binary stepwise logistic regression was used to determine whether vestibular efficacy - as defined by the participant’s SOT VEST sensory ratio measures - is a significant predictor of abnormal balance, independent of striatal monoaminergic terminal DVR or other confounders. Forward stepwise selection process was used with the VEST, VIS, SOM and PREF SOT sensory ratio measures (root mean square, RMS, frequency, and speed of center) of pressure (COP) excursions, participant's age, sex, body mass index, height, weight, LED, duration of PD, standardized cognitive assessment, and motor UPDRS scores, vibration sensitivity (biothesiometer and tuning fork), thalamic/cortical AChE hydrolysis rates and nigrostriatal dopaminergic denervation. Akaike Information Criterion (AIC) was chosen as the model entry criterion. Testing the resulting set of predictors for collinearity based on a correlation coefficient matrix did not show evidence of multicollinearity (all R's < 0.5).
Participants that fell in any of the SOT 1-4 conditions were excluded as non-vestibular sensory changes may be driving the postural deficits in the person (n = 2). Participants who fell on SOT 5 (n = 13) were excluded from the main stepwise logistic regression analysis where the missing VEST sensory ratio was the main independent variable but were included in a post-hoc 2x2 table analysis of binarized samples of PD patients with normal versus abnormal balance and normal versus abnormal vestibular efficacy (the latter based on NC normative data). Vestibular efficacy abnormality was computed by transforming all the PD patient values for COP RMS for VEST into control participant normalized z-scores, by subtracting the control mean and dividing by the control standard deviation. Any z-scores more than 1.649 standard deviations from the normal control mean were interpreted as abnormal vestibular efficacy. PD patients who fell during the completion of the SOT 5 condition were also defined as having abnormal vestibular efficacy. Abnormal balance was defined in the same manner as in the primary analysis. The resulting 2x2 table was analyzed using Fisher’s exact test. Statistical analysis was performed using R version 4.1.1, R Core Team (Vienna, Austria).
3. Results
3.1. Descriptive statistics
This cross-sectional study involved 135 participants (95 males, 40 females), mean age 65.7±8.8 (range 40-86 years). 106 of these participants were PD patients, and the other 29 were normal controls (NC; 13 males, 16 females), mean age 64.9±12.3 (range 40-84 years). The average Montreal Cognitive Assessment (MOCA) test score was 26 ± 2.3 (range 19-30) [10]. Average duration of motor symptoms observed among PD patients was 6.1±4.1 years (range 0.5-19). The average Hoehn and Yahr (HY) stage among patients was 2.4±0.6 (range 1-4.0) [8]. The mean Movement Disorder Society-revised Unified Parkinson Disease Rating Scale (UPDRS) in the medication “off” state after withholding of dopaminergic medication overnight was 37.1±14.7 (range 9.5-73.5) [9]. Abnormal balance was present in 73 out of 106 patients.
3.2. Group comparison
Both normal balance PD (B = 0.17, p = 0.0037) and NC (B = 1.3, p < 0.001) had greater dopaminergic innervation of the striatum relative to the abnormal balance PD group (table 1). Only the NC group showed greater cholinergic innervation of the cortex (B = 0.0066, p < 0.001), with no significant difference detected between normal and abnormal balance PD participants. No significant difference was found between either of the normal balance groups and abnormal balance PD group on cholinergic innervation of the thalamus.
Table 1.
Table of ANOVA planned contrasts, with abnormal balance PD as baseline, and the two normal balance groups as contrasts. Significance of individual ANOVA contrasts was only emphasized if the respective ANOVA model came out significant. Descriptive information for each of the predictors is given by group as (mean ± stdev).
| Baseline | Contrasts | |||
|---|---|---|---|---|
| Predictor | PD with abnormal clinical balance (HY ≥ 2.5; n=73) |
PD with normal clinical balance (HY < 2.5; n=33) |
Normal controls (n = 29) |
Model Significance |
| Age (yr.) | 67.3 ± 7.6 | −4.3 (63.0 ± 7.05) | −2.4 (64.9 ± 12.3) | F(2, 132) = 2.8736 P(>∣F∣) = 0.3848 |
| Weight (lbs.) | 186.4 ± 37.7 | +11.7 (198.2 ± 39.8) | −12.4 (174.01 ± 41.0) | F(2, 132) = 2.9656 P(>∣F∣) = 0.3848 |
| Height (in.) | 68.1 ± 3.2 | +1.0 (69.06 ± 4.08) | −1.4 (66.7 ± 3.8) | F(2, 132) = 3.4863 P(>∣F∣) = 0.2695 |
| BMI | 28.1 ± 4.7 | +1.07 (29.2 ± 5.2) | −0.8 (27.3 ± 4.6) | F(2, 132) = 1.2565 P(>∣F∣) = 0.8521 |
| Striatal VMAT2 DVR | 1.9 ± 0.3 | +0.2 * (2.03 ± 0.3) | +1.3 * (3.2 ± 0.3) | F(2, 129) = 234.2081 P(>∣F∣) < 0.0001 |
| Thalamic AChE hydrolysis rates | 0.05 ± 0.006 | −0.001 (0.05 ± 0.005) | +0.006 (0.06 ± 0.008) | F(2, 103) = 2.7964 P(>∣F∣) = 0.3848 |
| Cortical AChE hydrolysis rates | 0.02 ± 0.003 | +0.0008 (0.02 ± 0.002) | +0.007 * (0.03 ± 0.007) | F(2, 103) = 9.8273 P(>∣F∣) = 0.0014 |
| Vibration sensitivity (tuning fork) | 8.8 ± 4.1 | −1.04 (7.8 ± 3.2) | +0.2 (9.01 ± 2.9) | F(2, 130) = 1.0910 P(>∣F∣) = 0.8521 |
| Vibration sensitivity (biothesiometer) | 26.4 ± 11.6 | +3.04 (29.4 ± 11.4) | −1.4 (25.0 ± 11.02) | F(2, 130) = 1.2709 P(>∣F∣) = 0.8521 |
| Global cognition | −0.6 ± 1.06 | +0.6 * (0.03 ± 0.6) | +0.8 * (0.2 ± 0.4) | F(2, 131) = 12.1119 P(>∣F∣) = 0.0002 |
| MOCA | 25.7 ± 2.3 | +0.9 (26.7 ± 2.2) | +1.2 (26.9 ± 2.09) | F(2, 132) = 3.6033 P(>∣F∣) = 0.2695 |
| MDS UPDRS total motor score | 41.01 ± 14.1 | −12.6 * (28.4 ± 12.1) | −38.05 * (3.0 ± 3.08) | F(2, 132) = 102.9014 P(>∣F∣) < 0.0001 |
| Duration of PD (yr.) | 6.8 ± 4.2 | −2.4 * (4.4 ± 3.4) | F(1, 104) = 8.2813 P(>∣F∣) = 0.0486 | |
| LED | 767.9 ± 623.2 | −148.2 (619.7 ± 336.2) | F(1, 104) = 1.6428 P(>∣F∣) = 0.8111 | |
P(>∣t∣) < 0.05
Abbreviations: AChE, acetylcholinesterase; MDS-UPDRS, Movement Disorder Society-revised Unified Parkinson's Disease Rating Scale; MOCA, Montreal Cognitive Assessment; VMAT2, vesicular monoamine transporter type 2.
Several clinical variables also showed significant differences when comparing normal balance groups against impaired balance PD participants (table 1). Normalized global cognitive assessment scores were greater among both normal balance PD participants (B = 0.64, p < 0.001) and NC participants (B = 0.84, p < 0.001). Lower motor UPDRS scores were also observed among normal balance PD (B = −12.59, p < 0.001) and NC participants (B = −38.05, p < 0.001) relative to abnormal balance PD participants. Finally, when comparing normal balance PD and abnormal balance PD participants on duration of PD and LED, only duration of PD (B = −2.42, p = 0.0049) was significantly different between groups, with shorter duration of disease on average in the normal balance PD group.
Fisher’s exact test showed that the distribution of sex across our participant groups (table 2) was not equal (p = 0.0048; n = 135). The difference in distribution of sex across groups is primarily driven by difference between the two PD groups and the NC group. NC participants show an equal distribution of sex (female = 16, male = 13), while the PD group with normal balance (female = 7, male = 26) and the PD group with abnormal balance (female = 17, male = 56) are both predominantly male. The distribution of sex between the two PD groups appears roughly equal.
Table 2.
Frequency table of sex by participant group. Fisher’s exact test for count data was conducted on this table to test for differences in sex distribution across groups.
| Group | Sex | Total | |
|---|---|---|---|
| female | male | ||
| PD Abnormal Balance | 17 | 56 | 73 |
| PD Normal Balance | 7 | 26 | 33 |
| NC | 16 | 13 | 29 |
| Total | 40 | 95 | 135 |
Fisher’s exact test: p = 0.004808
3.3. Binary logistic regression of abnormal balance
After forward selection, the final model (total model χ2 = 42.9, p < 0.0000001) had 11 predictors, 7 of which were significant (table 3). Among clinical and demographic predictors, motor UPDRS score (β = 0.11, p = 0.004) and age (β = 0.33, p = 0.001) came out as significant positive predictors of abnormal balance, while standardized cognitive assessment score (β = −1.72, p = 0.006) and vibration sensitivity (β = −0.17, p = 0.001) came out as significant negative predictors. Thalamic AChE (β = 426.67, p = 0.003) was the only significant PET predictor of balance with cortical AChE (β = −349.89, p = 0.15) present in the model but not significant, and striatal DTBZ was excluded during the forward selection process. Only two significant SOT ratio predictors were present in the final model, with COP RMS of the vestibular ratio (β = 0.68, p = 0.002) predicting a greater likelihood of abnormal balance and speed of COP under the visual ratio (β = 0.68, p = 0.012) predicting a lesser likelihood of abnormal balance. Speed of COP under the somatosensory ratio and RMS of COP under the somatosensory ratios were included in the final model and approached statistical significance (table 3).
Table 3.
Results from the stepwise logistic regression model listing predictors of abnormal balance in the PD patients (total model χ2= 42.9, p < 0.0000001). The listed variables fulfilled the retention criteria to stay in the stepwise selection model. Abbreviations: COP, center of pressure; RMS, root mean square; SOM, somatosensory ratio; VEST, vestibular sensory ratio; VIS, visual sensory ratio.
| Predictor | β | OR | Std. Error | p |
|---|---|---|---|---|
| Intercept | −33.76 | 2.2e−15 | 10.23 | 0.000967 |
| Age (yr) | 0.33 | 1.39 | 0.10 | 0.000954 |
| MDS UPDRS total motor score | 0.11 | 1.12 | 0.04 | 0.003922 |
| Global cognition | −1.72 | 0.18 | 0.63 | 0.006188 |
| Vibration Sensitivity (biothesiometer) | −0.17 | 0.84 | 0.05 | 0.000848 |
| RMS COP VIS | 0.68 | 1.97 | 0.39 | 0.078397 |
| Speed COP VIS | −0.83 | 0.44 | 0.33 | 0.012290 |
| RMS COP SOM | −1.60 | 0.20 | 0.89 | 0.073581 |
| Speed COP SOM | −0.94 | 0.39 | 0.95 | 0.319741 |
| RMS COP VEST | 0.68 | 1.97 | 0.21 | 0.001571 |
| Thalamic AChE hydrolysis rates | 426.67 | 2.0e+185 | 144.51 | 0.003151 |
| Cortical AChE hydrolysis rates | −349.89 | 1.1e−152 | 240.93 | 0.146441 |
3.4. Post-hoc Chi-Square analysis to account for SOT 5 condition fallers.
Our main analysis was based on continuous SOT sensory ratio measures, but this excludes the subset of 13 PD patients who fell during the SOT 5 condition. Post-hoc contingency table (table 4), Fisher’s exact test was conducted to examine the association between vestibular efficacy and balance. Fisher’s exact test (p = 0.0002908; n = 106) showed a significant association between vestibular efficacy and balance. The distribution of normal versus abnormal balance among participants with normal vestibular efficacy is roughly equal (normal = 30, abnormal = 40). When considering participants with abnormal vestibular efficacy, a disproportional number of them also had abnormal balance (normal = 3, abnormal = 33).
Table 4.
2x2 table results for binarized balance (normal vs. abnormal) and vestibular efficacy (normal vs. abnormal) in the 106 PD patients. Patients who fell during the sensory organization test (SOT) condition 5 were categorized as having abnormal vestibular efficacy whereas abnormal vestibular efficacy in those who did not fall in this condition was based on deviation from normative data in the control group. Abnormal balance was defined as HY ≥ 2.5.
| Balance | Vestibular Efficacy | Total | |
|---|---|---|---|
| Normal | Abnormal | ||
| Normal | 30 | 3 | 33 |
| Abnormal | 40 | 33 | 73 |
| Total | 70 | 36 | 106 |
Fisher’s exact test: p = 0.0002908
4. Discussion
Postural imbalance is one of the major clinical symptoms of PD, but its underlying pathophysiology remains poorly understood. Effective postural control depends in part on integration of sensory information from proprioceptive, visual and vestibular functions. Sensory integration may be affected in patients with PD with abnormal balance [21]. PD may affect postural control, especially when a concurrent task is performed or when encountering a novel situation that may require an adjustment of the postural strategy. In particular, dynamic and sensory conflicting situations can be difficult for patients with PD with multisensory integration deficits [21]. Our findings show that vestibular efficacy is a determinant of abnormal balance in PD independent from nigrostriatal degeneration.
This model accounts for all six sensory conditions of the SOT task, allowing relatively specific isolation of the inputs from the vestibular, somatosensory and visual postural processing sensory systems. Increased COP sway surface of the vestibular SOT sensory ratio predicted impaired balance independent from the degree of nigrostriatal degeneration and was relatively more robust compared to visual and somatosensory postural control processing changes. This finding suggests that decreased vestibular efficacy during postural control is an independent component of abnormal balance in PD.
We previously have shown that decreased striatal dopamine function in normal aging does not impair the ability of the vestibular control system during SOT 5 indicating that this likely represents a non-dopaminergic function [6]. Our current findings show that processing of vestibular information is impaired in PD but - as predicted - appears to be independent from the degree of striatal dopaminergic losses. Though our group comparison showed that both PD with normal balance and controls have a greater level of dopaminergic integrity relative to PD with abnormal balance, dopaminergic innervation of the striatum was not retained in the final stepwise regression model, likely because it accounts for variance that overlaps strongly with motor UPDRS scores. Given that motor UPDRS scores are known to correlate strongly with dopaminergic system integrity in PD progression, all other significant predictors remaining in the model associate with abnormal balance independently from loss of dopaminergic innervation in PD. We also found that decreased cognition was a significant regressor in the final model. This is in keeping that patients with PD with more severe PIGD motor features tend to have also more severe cognitive problem that may in part may reflect a shared pathophysiology [30, 31].
Our finding of a contribution of vestibular efficacy to balance functions in PD augurs vestibular approaches to treat axial motor impairments in PD. For example, a study provided proof of concept that stochastic vestibular galvanic stimulation improved postural functions in PD 23. Along with our findings, these results suggest that vestibular sensory processing plays an important role in monitoring and improving postural functions in patients with PD.
Deficits in vestibular sensory processing may reflect pure vestibular (peripheral and/or central) deficits but could also result from downstream deficits in processing of vestibular sensory information in thalamic relay stations, such as the medial geniculate body or the vestibular temporal cortex. For example, we recently reported that vesicular acetylcholine transporter deficits in the medial geniculate body can be seen with impaired balance and gait functions in patients with PD [24]. We also previously reported that decreased functions of thalamic efferent of cholinergic pedunculopontine nucleus neurons may contribute to postural control in PD, likely by playing a role in the multimodal integration of sensory input information [11]. More recently, we reported on the role of vesicular acetylcholine transporter changes in the medial geniculate body associating with postural instability and gait difficulties in an independent population of PD persons and using a cholinergic PET ligand that reliably allows quantification of radioligand binding in small sized thalamic parcellations [32]. In accord with these prior findings, cholinergic innervation of the thalamus came out as the strongest predictor of abnormal balance in our binary stepwise regression analysis in the current study. Collectively, these findings point to the cholinergic thalamus as an important hub in a vestibular neural network that is altered in persons with PD and imbalance.
Our current data show that decreased vestibular efficacy in PD contributes to abnormal balance not only independent from nigrostriatal but also pendunculopontine nucleus-thalamic cholinergic degeneration. It should be noted, however, that cholinergic nerve terminals exist within the vestibular brainstem nuclei, such as the caudal medial vestibular nucleus [22], and that anti-cholinergic drugs may help to relieve symptoms of motion sickness [23]. Therefore, a cholinergic modulation of processing of vestibular sensory information cannot be excluded. Furthermore, our AChE PET imaging technique is limited for identifying cholinergic activity in anatomic small-sized nuclei, such as the vestibular nuclei.
We chose to use the HY to define abnormal balance over other motor assessments due to the HY being a solid predictor of disability and quality of life for PD patients, at least when performed by a single examiner. We had one examiner score the HY for each participant, which increases consistency. However, the use of HY to account for abnormal balance could be a limiting factor if we did not normalize for comorbidities that may impact balance, such as neuropathy. Past research has shown that peripheral neuropathy is significantly associated with increased PIGD in participants with and without peripheral neuropathy when matched for HY scores [25]. Our assessment of vibration sensitivity helps us account for any sensory neuropathies experienced by participants. The fact that vibration sensitivity came out as a significant predictor in our binary stepwise regression suggests that neuropathy does indeed play a role in abnormal balance. By virtue of its significance and inclusion in the model after stepwise regression, we can infer that the effect of vestibular impairment on abnormal balance is independent of any potential neuropathy experienced by the participant. Though balance was assessed in the dopaminergic ‘off’ state, vestibular efficacy testing was performed in the dopaminergic 'on' state. However, LED was not a significant regressor in our multivariate analysis.
Although we estimated the relative but more specific impact of vestibular sensory processing by using the vestibular sensory ratios, our study was limited in that we did not perform vestibular testing based on nystagmography or caloric testing to identify a vestibular peripheral or central deficit more precisely in the PD patients. Future studies should employ dedicated vestibular testing to confirm our findings. Another limitation of the study is that the retropulsion test is a test of reactive postural adjustment whereas the SOT measures are assessed under more steady state postural control tasks. All of these assessments will involve processing of vestibular information in order to maintain postural control but there will be differences in the relative involvement of visual and ankle proprioceptive sensory information processing. We also show that a more critical role is being played by vestibular postural control responses which is of key importance for the retropulsion test as we performed this test with eyes open and on stable and solid flooring (i.e., minimizing dependence on visual and ankle proprioceptive sensory processing).
The findings in this study show that vestibular efficacy is a significant determinant of abnormal balance observed in PD, in a manner that is independent of both cholinergic and dopaminergic system integrity, and independent of contributions of visual and somatosensory system. However, visual and somatosensory sensory processing systems also contribute to abnormal balance in PD, albeit to a more modest extent. To our knowledge, this study is among the first to investigate the relationship between sensory processing during postural control system efficacy and abnormal balance in PD while accounting for differences in dopaminergic and cholinergic innervation as determined by PET imaging.
Highlights.
All three sensory systems (vestibular, visual, somatosensory) contribute to balance impairment in Parkinson's disease (PD).
Efficacy of vestibular processing is a determinant of balance impairment in PD independent from nigrostriatal degeneration.
Cholinergic innervation of the thalamus predicts balance abnormality and may account for sensory integration deficits in PD.
Acknowledgements
The authors thank Christine Minderovic, Virginia Rogers, the PET technologists, cyclotron operators, and chemists, for their assistance.
Funding agencies:
This work was supported by the Department of Veterans Affairs (I01 RX001631); the Michael J. Fox Foundation; and the NIH [grant numbers P01 NS015655, P50 NS091856, R01 NS099535 and RO1 NS070856].
Abbreviations:
- AChE
Acetylcholinesterase
- COP
center of pressure
- DVR
Distribution volume ratio
- HY
Hoehn & Yahr stage
- PD
Parkinson disease
- PET
positron emission tomography
- RMS
root mean square
- SOT
sensory organization test
- SOM
somatosensory ratio
- VEST
vestibular sensory ratio
- VIS
visual sensory ratio
- VMAT2
vesicular monoamine transporter type 2
Footnotes
Relevant conflicts of interest/financial disclosures: The authors have no relevant financial or conflict of interest to disclose.
References
- [1].Smith PF, Vestibular Functions and Parkinson's Disease, Front Neurol 9 (2018) 1085. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Reichert WH, Doolittle J, McDowell FH, Vestibular dysfunction in Parkinson disease, Neurology 32(10) (1982) 1133–8. [DOI] [PubMed] [Google Scholar]
- [3].de Natale ER, Ginatempo F, Paulus KS, Manca A, Mercante B, Pes GM, Agnetti V, Tolu E, Deriu F, Paired neurophysiological and clinical study of the brainstem at different stages of Parkinson's Disease, Clin Neurophysiol 126(10) (2015) 1871–8. [DOI] [PubMed] [Google Scholar]
- [4].de Natale ER, Ginatempo F, Paulus KS, Pes GM, Manca A, Tolu E, Agnetti V, Deriu F, Abnormalities of vestibular-evoked myogenic potentials in idiopathic Parkinson's disease are associated with clinical evidence of brainstem involvement, Neurol Sci 36(6) (2015) 995–1001. [DOI] [PubMed] [Google Scholar]
- [5].Pastor MA, Day BL, Marsden CD, Vestibular induced postural responses in Parkinson's disease, Brain 116 ( Pt 5) (1993) 1177–90. [DOI] [PubMed] [Google Scholar]
- [6].Cham R, Perera S, Studenski SA, Bohnen NI, Striatal dopamine denervation and sensory integration for balance in middle-aged and older adults, Gait Posture 26(4) (2007) 516–25. [DOI] [PubMed] [Google Scholar]
- [7].Hughes AJ, Daniel SE, Kilford L, Lees AJ, Accuracy of clinical diagnosis of idiopathic Parkinson's disease: a clinico-pathological study of 100 cases, J Neurol Neurosurg Psychiatry 55(3) (1992) 181–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Goetz CG, Poewe W, Rascol O, Sampaio C, Stebbins GT, Counsell C, Giladi N, Holloway RG, Moore CG, Wenning GK, Yahr MD, Seidl L, Movement Disorder Society Task Force report on the Hoehn and Yahr staging scale: status and recommendations, Mov Disord 19(9) (2004) 1020–8. [DOI] [PubMed] [Google Scholar]
- [9].Emre M, Aarsland D, Brown R, Burn DJ, Duyckaerts C, Mizuno Y, Broe GA, Cummings J, Dickson DW, Gauthier S, Goldman J, Goetz C, Korczyn A, Lees A, Levy R, Litvan I, McKeith I, Olanow W, Poewe W, Quinn N, Sampaio C, Tolosa E, Dubois B, Clinical diagnostic criteria for dementia associated with Parkinson's disease, Mov Disord 22(12) (2007) 1689–707. [DOI] [PubMed] [Google Scholar]
- [10].Nasreddine ZS, Phillips NA, Bedirian V, Charbonneau S, Whitehead V, Collin I, Cummings JL, Chertkow H, The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment, J Am Geriatr Soc 53(4) (2005) 695–9. [DOI] [PubMed] [Google Scholar]
- [11].Muller ML, Albin RL, Kotagal V, Koeppe RA, Scott PJ, Frey KA, Bohnen NI, Thalamic cholinergic innervation and postural sensory integration function in Parkinson's disease, Brain 136(Pt 11) (2013) 3282–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].Prieto TE, Myklebust JB, Hoffmann RG, Lovett EG, Myklebust BM, Measures of postural steadiness: differences between healthy young and elderly adults, IEEE Trans Biomed Eng 43(9) (1996) 956–66. [DOI] [PubMed] [Google Scholar]
- [13].Vanicek N, King SA, Gohil R, Chetter IC, Coughlin PA, Computerized dynamic posturography for postural control assessment in patients with intermittent claudication, J Vis Exp (82) (2013) e51077. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [14].Pletcher ER, Williams VJ, Abt JP, Morgan PM, Parr JJ, Wohleber MF, Lovalekar M, Sell TC, Normative Data for the NeuroCom Sensory Organization Test in US Military Special Operations Forces, J Athl Train 52(2) (2017) 129–136. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [15].Bohnen NI, Muller ML, Kotagal V, Koeppe RA, Kilbourn MR, Gilman S, Albin RL, Frey KA, Heterogeneity of cholinergic denervation in Parkinson's disease without dementia, J Cereb Blood Flow Metab 32(8) (2012) 1609–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [16].Shao X, Hoareau R, Hockley BG, Tluczek LJ, Henderson BD, Padgett HC, Scott PJ, Highlighting the Versatility of the Tracerlab Synthesis Modules. Part 1: Fully Automated Production of [18F]Labelled Radiopharmaceuticals using a Tracerlab FXFN, J Labelled Comp Radiopharm 54(6) (2011) 292–307. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [17].Koeppe RA, Frey KA, Kume A, Albin R, Kilbourn MR, Kuhl DE, Equilibrium versus compartmental analysis for assessment of the vesicular monoamine transporter using (+)-alpha-[11C]dihydrotetrabenazine (DTBZ) and positron emission tomography, J Cereb Blood Flow Metab 17(9) (1997) 919–31. [DOI] [PubMed] [Google Scholar]
- [18].Minoshima S, Koeppe RA, Fessler JA, Mintun MA, Berger KL, Taylor SF, Kuhl DE, Integrated and automated data analysis method for neuronal activation studying using O15 water PET., in: Uemura K, Lassen NA, Jones T, Kanno I (Eds.), Quantification of brain function to tracer kinetics and image analysis in brain PET., Excerpta Medica, Tokyo, 1993, pp. 409–418. [Google Scholar]
- [19].Logan J, Fowler JS, Volkow ND, Ding YS, Wang GJ, Alexoff DL, A strategy for removing the bias in the graphical analysis method., J Cereb Blood Flow Metab 21 (2001) 307–320. [DOI] [PubMed] [Google Scholar]
- [20].Nagatsuka S, Fukushi K, Shinotoh H, Namba H, Iyo M, Tanaka N, Aotsuka A, Ota T, Tanada S, Irie T, Kinetic analysis of [(11)C]MP4A using a high-radioactivity brain region that represents an integrated input function for measurement of cerebral acetylcholinesterase activity without arterial blood sampling, J Cereb Blood Flow Metab 21(11) (2001) 1354–66. [DOI] [PubMed] [Google Scholar]
- [21].Colnat-Coulbois S, Gauchard GC, Maillard L, Barroche G, Vespignani H, Auque J, Perrin PP, Management of postural sensory conflict and dynamic balance control in late-stage Parkinson's disease, Neuroscience 193 (2011) 363–9. [DOI] [PubMed] [Google Scholar]
- [22].Barmack NH, Central vestibular system: vestibular nuclei and posterior cerebellum, Brain Res Bull 60(5-6) (2003) 511–41. [DOI] [PubMed] [Google Scholar]
- [23].Hain TC, Uddin M, Pharmacological treatment of vertigo, CNS Drugs 17(2) (2003) 85–100. [DOI] [PubMed] [Google Scholar]
- [24].Bohnen NI, Kanel P, Koeppe RA, Sanchez-Catasus CA, Frey KA, Scott P, Constantine GM, Albin RL, Muller MLTM, Regional cerebral cholinergic nerve terminal integrity and cardinal motor features in Parkinson’s disease, Brain Communications 3(2) (2021) fcab109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [25].Beaulieu ML, Muller MLTM, Bohnen NI, Peripheral neuropathy is associated with more frequent falls in Parkinson’s disease, Parkinsonism and Related Disorders 54(2018) 46–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [26].Ramaker C, Marinus J, Stiggelbout AM, van Hilten BJ, Systematic Evaluation of Rating Scales for Impairment and Disability in Parkinson’s Disease, Movement Disorders 17(2002) 867–876. [DOI] [PubMed] [Google Scholar]
- [27].Chaudhry H, Bukiet B, Ji Z, Findley T, Measurement of Balance in Computer Posturography: Comparison of Methods - A Brief Review, Journal of Bodywork & Movement Therapies 15 (2011) 82–91. [DOI] [PubMed] [Google Scholar]
- [28].Cohen HS, Kimball KT, Usefulness of Some Current Balance Tests for Identifying Individuals with Disequilibrium Due to Vestibular Impairments, Journal of Vestibular Research 18 (2008) 295–303. [PMC free article] [PubMed] [Google Scholar]
- [29].Holm S A Simple Sequentially Rejective Multiple Test Procedure. Scandinavian Journal of Statistics, 6(2), (1979) 65–70. [Google Scholar]
- [30].Alves G, Larsen JP, Emre M, Wentzel-Larsen T, Aarsland D. Changes in motor subtype and risk for incident dementia in Parkinson's disease. Mov Disord. 2006. Aug;21(8):1123–30. [DOI] [PubMed] [Google Scholar]
- [31].Zuo LJ, Piao YS, Li LX, Yu SY, Guo P, Hu Y, Lian TH, Wang RD, Yu QJ, Jin Z, Wang YJ, Wang XM, Chan P, Chen SD, Wang YJ, Zhang W. Phenotype of postural instability/gait difficulty in Parkinson disease: relevance to cognitive impairment and mechanism relating pathological proteins and neurotransmitters. Sci Rep. 2017. Mar 23;7:44872. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [32].Bohnen NI, Kanel P, Roytman S, Scott P, Koeppe RA, Albin RL, Kerber KA, & Müller M Cholinergic brain network deficits associated with vestibular sensory conflict deficits in Parkinson's disease: correlation with postural and gait deficits. Journal of neural transmission, (2022). Advance online publication. [DOI] [PMC free article] [PubMed] [Google Scholar]