Cerebellar White Matter Damage Is Associated With Postural Sway Deficits in People With Multiple Sclerosis

Archives of Physical Medicine and Rehabilitation journal homepage: www.archives-pmr.org Archives of Physical Medicine and Rehabilitation 2019;-:-------

ORIGINAL RESEARCH

Cerebellar White Matter Damage Is Associated With Postural Sway Deficits in People With Multiple Sclerosis Geetanjali Gera, PhD, PT,a Brett W. Fling, PhD,b Fay B. Horak, PhD, PTc From the aDepartment of Physical Therapy, University of Kentucky, Lexington, Kentucky; bDepartment of Health and Exercise Science, Colorado State University, Fort Collins, Colorado; and cDepartment of Neurology, School of Medicine, Oregon Health & Science University, Portland, Oregon.

Abstract Objective: To assess how postural sway deficits during eyes open and closed relate to the integrity of cerebellar peduncles in individuals with multiple sclerosis (MS). Design: Cross-sectional study. Setting: Laboratory based setting. Participants: Twenty-nine adults with MS (Expanded Disability Status Scale: 2-4) and 15 adults without MS were recruited (NZ44). Inclusion criteria for all participants were ability to maintain balance independently by standing on toes for 3 seconds, and no known biomechanical conditions affecting balance. Interventions: Not applicable. Main Outcome Measures: Postural sway using body-worn, inertial sensors during quiet standing, integrity of cerebellar peduncles quantified using diffusion-tensor imaging and clinical assessment scales for ataxia and balance. Results: Radial diffusivity of the inferior cerebellar peduncle was related to postural sway measures during both eyes open and closed. In contrast, radial diffusivity of the superior cerebellar peduncle was related to postural sway only in stance with eyes open. Conclusions: The inferior cerebellar peduncle, which carries somatosensory information to the cerebellum, contributes to control of standing balance with or without visual inputs, consistent with the high dependence on somatosensory information for posture. The superior cerebellar peduncle, which carries cortical information to the cerebellum, contributes to control of standing posture only when vision is available. Radial diffusivity of the inferior cerebellar peduncle was related to reactive balance control, whereas radial diffusivity of the superior cerebellar peduncle was related to the kinetic component of the ataxia rating scale. Archives of Physical Medicine and Rehabilitation 2019;-:------ª 2019 Published by Elsevier Inc. on behalf of the American Congress of Rehabilitation Medicine

Balance impairments are a significant problem in people with multiple sclerosis (PwMS), contributing to high incidence of falls and reduced quality of life.1,2 Qualitative assessment of postural sway for balance control during quiet standing with eyes open and closed is frequently used to identify balance deficits due to either cerebellar or Supported by the National Multiple Sclerosis Society (RG-5273; Fling; FG 2058-A-1 Gera; MB-0027; Horak) and the Medical Research Foundation of Oregon (Fling, Gera). Disclosures: F. Horak has a significant financial interest in APDM, a company that may have a commercial interest in the results of this research and technology. This potential conflict of interest has been reviewed and managed by Oregon Health & Science University and the Integrity Program Oversight Council. F. Horak discloses financial relationships with Medtronics, Takeda, Adamus, Sanofi, Neuropore, and Biogen outside the submitted work. The other authors have nothing to disclose.

proprioceptive damage. Postural sway deficits in PwMS during standing with eyes open and closed have been quantified with laboratory measures such as center of pressure path derived from force platform or trunk sway trajectory based on motion analysis, thereby limiting the use of this measure by clinicians.3 Recently, body-worn inertial sensors on the pelvis have been used to quantify postural sway deficits during quiet standing, an approach available to clinicians.4,5 Integrating postural sway measured via inertial-sensor linear accelerations in the anterior-posterior (AP) and mediolateral directions, quantifies jerkiness of sway, thought to be related to active, central nervous system postural corrections that we postulated would be excessive in PwMS.6

0003-9993/19/$36 - see front matter ª 2019 Published by Elsevier Inc. on behalf of the American Congress of Rehabilitation Medicine https://doi.org/10.1016/j.apmr.2019.07.011

2

G. Gera et al

Postural sway is controlled by feedback from vision, vestibular, and somatosensory systems, with the largest reliance on somatosensory inputs entering the cerebellum via the inferior peduncle. Visual and vestibular information to the cerebellum is carried by the middle peduncle. Postural control also relies upon a feedforward internal model of the body and environment that likely is updated by cerebral inputs to the cerebellum via the superior peduncle.7 However, it is unclear whether abnormal postural sway in PwMS is due to disrupted somatosensory conduction from the spinal cord to the cerebellum via the inferior peduncle or from the cerebellum to the cerebral cortex via the superior peduncle. Studies investigating the neural mechanisms underlying associations between postural sway during eyes open and closed conditions have demonstrated atrophy of cerebellum and spinal cord structures in PwMS.3,8 However, atrophy is an end-stage phenomenon. Abnormalities in the brain and spinal cord pathways can occur even during the early stage of multiple sclerosis (MS), before atrophy becomes evident. Diffusion-tensor imaging (DTI) can measure these early pathologic changes by quantifying the integrity of white matter pathways. The cerebellar peduncles, that is, superior, middle, and inferior, are the white matter tracts to and from the cerebellum and therefore critical components of the neural network underpinning postural control. Integrity of the cerebellar peduncles can be affected in PwMS.3,8,9 In fact, deficits in the superior and middle cerebellar peduncles have been shown to be associated with increased postural sway when standing with eyes open in PwMS.3 However, the relationship between postural sway deficits during standing with eyes closed and cerebellar white matter tract integrity remains uninvestigated. The inferior, but not superior or middle, cerebellar peduncles carry afferent proprioceptive information via dorsal spinocerebellar tracts. Thus, we hypothesized that deficits in the inferior cerebellar peduncles would be specifically related to the postural sway deficits for standing with eyes closed. This is the first study to assess nontraditional measures of postural control using portable, inertial-sensor technology to (1) quantify postural sway during eyes open and closed standing conditions in PwMS; and (2) assess how postural sway deficits relate to the integrity of all 3 cerebellar white matter tracts. To evaluate the deficits in postural sway due to the involvement of cerebellum, we divided our cohort based on their scores on clinical ataxia measures.

Methods Participants Twenty-nine PwMS and 15 age- and sex-matched adults without MS were recruited. Inclusion criteria for all participants were (1)

List of Abbreviations: AP DTI EDSS FA ICARS Mini-BESTest MS PwMS RD

anterior-posterior diffusion-tensor imaging Expanded Disability Status Scale fractional anisotropy International Cooperative Ataxia Rating Scale Miniature Balance Evaluation Systems Test multiple sclerosis people with multiple sclerosis radial diffusivity

Expanded Disability Status Scale (EDSS)
4; (2) ability to maintain balance independently by standing on toes for 3 seconds; and (3) no known biomechanical conditions affecting balance. Exclusion criteria for PwMS and controls were: (1) co-existing conditions that can mimic MS (eg, lupus or fibromyalgia); (2) contraindications to undergo magnetic resonance imaging; or (3) additional conditions that may affect gait or balance (eg, arthritis, joint replacement). Of the 29 PwMS and 15 control subjects without MS, postural data from 2 control subjects were excluded due to inability to complete the protocol (nZ2). Data was also excluded from 4 PwMS because of disease type, that is, primary progressive MS (nZ2), technical issues (nZ1), or incomplete protocol (nZ1). Characteristics of remaining 25 PwMS and 13 control subjects are described in table 1. Three of the 25PwMS were not able to complete eyes closed testing.

Clinical assessment EDSS was used to determine the disease severity. EDSS scores for the PwMS ranged between 2 to 4 (table 1). International Cooperative Ataxia Rating Scale (ICARS) was used as a clinical scale to assess the extent of ataxia.10 ICARS has 4 subcomponents: Posture and gait, Kinetic, Speech, and Oculomotor disorders. Dynamic gait and balance were assessed in our participants using the Miniature Balance Evaluation Systems Test (Mini-BESTest), which has 4 subcomponents: Anticipatory, Reactive, Sensory, and Dynamic.11

Neuroimaging: DTI Image acquisition Participants were scanned on a 3.0T Magnetom Tim Trio scannera with a 12-channel head coil at the Oregon Health & Science University’s Advanced Imaging Research Center. One highresolution T1-weighted magnetization prepared rapid gradient echo sequence (orientation, sagittal; echo time, 3.58ms; repetition time, 2300ms; 256  256 matrix; resolution, 1.0  1.0  1.1mm; total scan time, 9min 14s) was acquired. A whole-brain echoplanar imaging sequence was also used (repetition time, 9100ms; echo time, 88ms; field of view, 240mm2; b value, 1000s/mm2; isotropic voxel dimensions, 2mm3). Images were sensitized for diffusion along 90 different directions with b value of 1000 s/mm2. For every 36 diffusion-weighted images, a nonediffusion-weighted image (bZ0s/mm2) was acquired (3 total). A static magnetic field map was also acquired using the same parameters as the diffusionweighted sequence. DTI analysis Diffusion data were processed using the tools implemented in FMRIB Software Library version 5.0.b The 3 raw data sets were first corrected for eddy current distortions and motion artifacts using FMRIB’s diffusion toolbox version 1.0,b then averaged to improve signal-to-noise ratio12 and subsequently skull-stripped (using FMRIB’s brain extraction toolb). Using the averaged images with bZ0 and bZ1000 s/mm2, the diffusion tensor was calculated. Diagonalization of the diffusion tensor yields the eigenvalues (l1, l2, and l3) as well as the eigenvectors that define the predominant diffusion direction. Using this approach, several measures have been developed and used to quantify white matter microstructural integrity. The most commonly used measures in PwMS are fractional anisotropy (FA) and radial diffusivity (RD). FA is a www.archives-pmr.org

Postural sway deficits multiple sclerosis Table 1

3

Subjects demographics and clinical characteristics

Clinical Score

Control nZ13

Mild-Ataxia MS nZ14

Moderate-Ataxia MS nZ11

Age (y), mean  SD M/F Disease duration (y), mean  SD Clinical course ICARS total, mean  SD ICARS (posture and gait), mean  SD ICARS (kinetic), mean  SD Mini-BESTest, mean  SD EDSS, mean  SD

49.013.26 3/10 N/A N/A N/A N/A N/A 26.01.8 N/A

48.611.5 2/12 11.06.6 2SP/12RR 5.12.7 2.62.2 1.61.8 24.23.5 3.070.85

44.08.4 3/8 12.88.4 1SP/1PR/9RR 22.98.2 7.33.2 14.26.5 19.62.9 3.910.2

NOTE. The ICARS has a maximum score of 100. The Mini-BESTest has a maximum score of 28. Abbreviations: F, female; M, male; N/A, not applicable; PR, progressive-relapsing multiple sclerosis; RR, relapsing-remitting multiple sclerosis; SP, secondary progressive multiple sclerosis.

rotationally invariant index that ranges from 0 (isotropic) to 1 (anisotropic). Therefore, higher FA values can be interpreted as reflecting higher white matter integrity.13 RD represents diffusion along secondary and tertiary axes where higher values represent poor structural integrity. Recent work has suggested RD to be specifically related to demyelination, a common pathology in MS.14,15 An a priori region of interest approach was selected to identify relationships between postural sway and the cerebellar peduncles white matter integrity using the methodology previously described by Pijnenburg et al16 (fig 1). Briefly, diffusion data were coregistered to Montreal Neurological Institute space and the Johns Hopkins University diffusion magnetic resonance imagingebased white matter atlas was used to identify the superior, middle, and inferior cerebellar peduncles within each hemisphere, respectively.

Postural control data recording An instrumented test of postural sway was administered during quiet stance under eyes open and eyes closed conditions while standing on a firm surface with the feet a standard distance apart (2.5cm between heels and 5cm between the base of the halluxes) via a template. A single, body-worn inertial Opal sensorc comprising tri-axial accelerometers, tri-axial gyroscopes, and a magnetometer, was used to provide objective measures of postural sway.6,17 The sensor was positioned with Velcrod belt on lower trunk, near the body center of mass at the L5 vertebra and data

were streamed to a laptop and automatically analyzed with MobilityLabc.

Postural control data analyses Balance objective measures were automatically derived from acceleration and angular velocity signals using the APDM Mobility Lab software.c We present here the postural sway measures from 4 identified domains of sway: (1) area; (2) jerk; (3) path length; and (4) frequency95. Postural sway area was computed as the area spanned by the acceleration signal per unit of time (m2/s5). Jerk measures jerkiness of the sway and is a time derivative of acceleration (m2/s5). Path length is the length of acceleration trajectory (m/s2). Frequency95 is the 95% power frequency, frequency below which the 95% of power of acceleration signal is present (Hz). We assessed both the AP and mediolateral components of postural sway, that is, path length, jerk, and frequency of sway.6

Statistical analyses We operationally divided the MS group into mild ataxia (nZ14) and moderate ataxia (nZ11) based on the ICARS scores of less than 10 and greater than 10 respectively. We performed the following statistical analyses: (1) a comparison of the 4 postural sway measures among the control and MS (mild-ataxia and moderate-ataxia) groups was evaluated with 3-way analysis of

Fig 1 Visualization of superior, middle and inferior cerebellar peduncles. Microstructural integrity of white matter tracts was quantified by RD of the peduncles as described by Pijnenburg et al.16

www.archives-pmr.org

4

G. Gera et al

variance; (2) assessment of the integrity of cerebellar white matter tracts for each control and MS group was also conducted with 3way analysis of variance among the control and MS (mild-ataxia and moderate-ataxia) groups for RD and FA values of superior, middle, and inferior cerebellar peduncles; and (3) associations between cerebellar function and postural sway measures were evaluated with Spearman correlation coefficients between cerebellar white matter tract integrity (RD and FA) and postural sway measures.

Results Assessment of postural sway Postural sway measures were significantly different between control and the moderate-ataxia MS group and between the mildand moderate-ataxia MS groups (table 2). Postural sway area, path

Table 2

length AP, and jerk AP differentiated the moderate-ataxia MS group from the mild-ataxia MS and control groups for both eyes open and closed conditions (see table 2). However, postural sway measures did not differ between the control and mild-ataxia MS groups, both for eyes open and closed conditions.

Integrity of cerebellar white matter tracts Figure 2 compares the structural integrity of cerebellar white matter tracts for the superior, middle, and inferior cerebellar peduncles among the 3 groups (control vs mild-ataxia MS vs moderate-ataxia MS, see fig 2). The moderate-ataxia MS group showed higher RD values for the superior cerebellar peduncle (P<.05) and inferior cerebellar peduncle (P<.05) than the control group and this difference approached significance for the middle cerebellar peduncle (PZ.07). The moderate-ataxia MS group also had higher radial diffusivity (RD) of superior (P<.05) but not for middle (PZ.27) and inferior (PZ.42) cerebellar peduncles, compared to the

Assessment of postural sway measures of disease severity and ataxia for control and multiple sclerosis groups

Postural Sway Measures 2

EO Control nZ13 MSmild nZ14 MSmod nZ11 Mean  SD

Group

EC Control nZ13 MSmild nZ13 MSmod nZ8 Mean  SD

Group

5

Area (m /s ) Control MSmild MSmod Jerk AP (m2/s5) Control MSmild MSmod Jerk ML (m2/s5) Control MSmild MSmod Path Length AP (m/s2) Control MSmild MSmod Path Length ML (m/s2) Control MSmild MSmod Frequency95 AP (Hz) Control MSmild MSmod Frequency95 ML (Hz) Control MSmild MSmod

0.0060.004 0.0090.005 0.0380.029

F value P

13.62 <.001*,y

0.0140.010 0.0220.016 0.0650.052

F value P

9.27 <.001*,y

1.4160.265 1.3260.347 1.8030.553

F value P

4.87 <.05 *,y

1.3690.375 1.4270.298 1.8860.651

F value P

3.96 <.05 *,y

1.5280.495 1.5160.321 1.8510.642

F value P

1.76 .19

1.7370.571 1.5810.248 1.9110.669

F value P

1.09 .35

4.3841.415 5.5541.876 13.2297.101

F value P

16.45 <.001*,y

6.6023.118 8.2042.847 16.9848.615

F value P

12.22 <.001*,y

5.0661.799 6.2501.704 13.8587.316

F value P

15.22 <.001*,y

7.8043.928 9.4364.342 19.58011.780

F value P

8.43 <.01*,y

1.9020.351 1.6020.457 2.2920.581

F value P

6.78 <.01*,y

1.9390.347 1.7970.461 2.1850.752

F value P

1.46 .25

1.6990.574 1.8350.562 2.3170.687

F value P

3.37 <.05*

1.7710.387 1.8410.350 2.2140.692

F value P

2.46 .10*

NOTE. Postural sway was assessed during eyes open and closed conditions in quiet standing. Group differences between control and multiple sclerosis mild were not significant. Abbreviations: EC, eye closed; EO, eyes open; ML, mediolateral; MS, multiple sclerosis; MSmild, multiple sclerosis mild; MSmod, multiple sclerosis moderate; SD, standard deviation. * Represents group differences between control and multiple sclerosis moderate P<.05. y Represents group differences between multiple sclerosis mild and multiple sclerosis moderate P<.05.

www.archives-pmr.org

Postural sway deficits multiple sclerosis Control

Radial diffusivity

1.2

Mild-ataxia MS

5 Moderate-ataxia MS

Table 4 Associations of radial diffusivity of cerebellar peduncles with clinical measures of ataxia and balance

*

1.0 *

0.8

Clinical Measure

0.6

0.4 0.2 0.0 Superior

Middle Cerebellar peduncles

Inferior

Fig 2 Assessment of the integrity of cerebellar peduncles for control and MS groups. RD of the cerebellar peduncles was higher (worse) for moderate MS as compared with the mild-ataxia MS and control groups.

mild-ataxia MS group. In contrast, RD of the 3 cerebellar peduncles did not differ between the mild-ataxia MS and the control groups (superior, PZ.94; middle, PZ.27; inferior, PZ.42). Only the moderate-ataxia MS group had lower FA values for the middle cerebellar peduncle (P<.05) compared to the control group. Other FA values did not differ among groups for superior, middle, and inferior peduncles (P>.08).

Relationships between cerebellar function and postural sway measures

ICARS total ICARS kinetic ICARS posture gait Mini-BESTest total Mini-BESTest anticipatory Mini-BESTest reactive Mini-BESTest sensory Mini-BESTest dynamic

SCP RD r, P Value

MCP RD r, P Value

ICP RD r, P Value

0.47, <.05* 0.57, <.01* 0.28, .19 -0.15, .48 -0.16, .46

0.19, .36 0.17, .43 0.29, .18 -0.28, .19 -0.30, .16

0.39, .06 0.40, .052 0.30, .15 -0.33, .11 -0.24, .26

-0.02, .91 -0.08, .72 -0.16, .46

-0.40, .05* 0.01, .98 -0.20, .34

-0.48, <.05* -0.11, .60 -0.25, .25

Abbreviations: ICP, inferior cerebellar peduncle; MCP, middle cerebellar peduncle; r, correlation coefficient; RD, radial diffusivity; SCP, superior cerebellar peduncle. * Statistically significant values.

peduncle approached significance with ICARS total and kinetic subcomponents. RD of the middle cerebellar peduncle was not related to ICARS total or kinetic subcomponents of ICARs. RD of the 3 peduncles was not related to the Mini-BESTest total scores. However, RD of the middle and inferior cerebellar peduncle was related to the reactive portion of the Mini-BESTest.

Discussion

RD of the superior cerebellar peduncle was related to postural sway measures only during the eyes open condition and not during the eyes closed condition (table 3). In contrast, RD of the inferior cerebellar peduncle was related to path length AP, jerk AP, and frequency 95 AP during both eyes open and closed stance. The relationship between the superior or inferior cerebellar white matter tract integrity (RD) was strongest with the jerk AP. FA of the peduncles was not related to the postural sway measures.

Relationship of cerebellar white matter tract integrity and ataxia and balance measures RD of the superior cerebellar peduncle was related to ICARS total and kinetic subcomponent (table 4). RD of the inferior cerebellar

This is the first study to differentiate quantitative measures of balance control with vision vs without vision related to cerebellar peduncles integrity as well as with clinical measures of ataxia and balance in people with moderate, relapsing-remitting MS. This study first showed that quantitative, objective measures of postural sway obtained during standing with feet apart, eyes open and closed with a body-worn inertial sensor can identify disrupted balance control in subjects with moderate MS compared to agematched control subjects. Higher derivatives of postural sway, that is, jerk, not often reported in the literature, was related with deficits in cerebellar white matter tract integrity to a greater extent than routinely reported sway measures, like sway path length. Second, this study showed that reduced integrity of superior and inferior cerebellar peduncle white matter contributes to the deficits

Table 3 Associations between radial diffusivity of cerebellar peduncles and postural sway measures during eyes open and closed conditions in quiet standing SCP RD

MCP RD

ICP RD

Postural Sway

EO r, P Value

EC r, P Value

EO r, P Value

EC r, P Value

EO r, P Value

EC r, P Value

Area Jerk AP Jerk ML Path Length AP Path Length ML Frequency95 AP Frequency95 ML

0.36, 0.49, 0.32, 0.40, 0.36, 0.43, 0.24,

0.20, .38 0.27, .24 -0.02, .92 0.26, .25 0.21, .36 0.27, .23 0.23, .31

0.12, 0.12, 0.09, 0.26, 0.02, 0.22, 0.19,

0.06, .78 0.35, .12 0.19, .40 0.08, .70 -0.09, .66 0.42, <.05* 0.26, .21

0.28, 0.42, 0.22, 0.41, 0.23, 0.34, 0.07,

0.28,0.21 0.52, <.05* 0.34, .13 0.40, .07 0.32, .15 0.48, <.05* 0.39, .08

.08 <.05* Z.11 <.05* .08 <.05* .24

.56 .57 .97 .31 .92 .33 .41

.17 <.05* .29 <.05* .26 .09 .75

Abbreviations: EC, eyes closed; EO, eyes open; ICP, inferior cerebellar peduncle; MCP, middle cerebellar peduncle; ML, mediolateral; r, correlation coefficient; RD, radial diffusivity; SCP, superior cerebellar peduncle. * Statistically significant values.

www.archives-pmr.org

6 in postural sway when vision is available (eyes open). In contrast, reduced integrity of the inferior cerebellar peduncle contributes to postural sway deficits when vision is not available (eyes closed). Third, this study showed reduced integrity of superior cerebellar peduncle white matter is reflected in the clinical measures of ataxia, that is, ICARS total and, specifically, the kinetic component of the ICARS assessment. In contrast, the reduced integrity of the middle and inferior cerebellar component contributes to the reactive balance component of the Mini-BESTest. Although our cohort was less impaired (EDSS: 2-4) than previous studies reporting deficits in the integrity of cerebellar peduncles and included only the relapsing-remitting type of MS, we did observe DTI deficits in the integrity of superior and inferior cerebellar peduncles for moderately ataxic PwMS compared to control individuals.3,8,9 In contrast, Anderson et al8 did not observe deficits in the integrity of superior and middle cerebellar peduncles for relapsing-remitting MS but did find reduced integrity of cerebellar peduncles for primary progressive MS. Prosperini et al3 reported deficits in the integrity of all 3 cerebellar peduncles for PwMS; however, the individuals included in their study were more severely affected and included some with a primary progressive course. Why does reduced integrity of superior and inferior cerebellar peduncles contribute to the deficits in postural sway when vision is available (eyes open), whereas reduced integrity of the inferior cerebellar peduncle affects postural sway when vision is not available (eyes closed)? Standing with eyes closed increases the reliance on proprioceptive information, thus explaining the importance of integrity of the inferior cerebellar peduncles that carry proprioceptive information via spinocerebellar pathways.18 Difficulty controlling postural sway with eyes closed is likely due to difficulty integrating proprioceptive information in the cerebellum to maintain balance. Our finding of the relationship between reduced integrity of the inferior cerebellar peduncle with postural sway in the eyes closed condition is inline with a study showing postural sway deficits in the eyes closed condition associated with atrophy in upper cervical spine area.3 However, gray matter atrophy is an end-stage phenomenon,3,9 which might go unnoticed in individuals who are mildly affected in early stages of the disease. In contrast, quantification of integrity of cerebellar white matter tracts with DTI has proven to be sensitive in differentiating clinically-impaired individuals with MS from those who did not have postural sway deficits.9 Thus, integrity of inferior cerebellar peduncle can potentially replace quantification of spinal cord deficits to investigate spinocerebellar pathways. The cerebellum, together with the vestibular nuclei and temporal-parietal junction are involved in integrating multisensory information from proprioceptive, visual and vestibular sources to control postural sway under changing sensory conditions.18 Our result showing postural sway measures during eyes open to be related to the reduced integrity of superior cerebellar peduncle is in agreement with a study showing that center of pressure path length during standing with eyes open was related to the reduced integrity of superior cerebellar peduncle.3 Unlike Prosperini et al,3 however, we did not find a relationship between structural integrity of the middle cerebellar peduncle and center of pressure path during eyes open in standing. It is conceivable that this difference in results is related to the lesser severity of our subjects. In

G. Gera et al addition, jerk (not previously reported) was better correlated with the reduced integrity of the superior and inferior cerebellar peduncles for eyes open and close conditions, respectively than the path length or area of sway displacement. Integrity of the inferior cerebellar peduncle, which carry afferent proprioceptive information via dorsal spinocerebellar tracts, was related to the reactive postural component of the MiniBESTest. The postural reaction section of the Mini-BESTest assesses postural responses to external perturbations, both with feet in place and compensatory stepping responses.11,19 When postural responses to external perturbations have been quantified with electromyogram or surface forces, their latencies are very long and related to delays in proprioceptive spinal afferent pathways (with somatosensory evoked potentials) in PwMS.20 Thus, the reactive portion of the Mini-BESTest may be used to assess deficits in the inferior cerebellar peduncle, although we cannot rule out other central nervous system contributions, as well as peripheral neuropathy.21 Although we did not observe a relationship between the components of Mini-BESTest balance assessment with superior cerebellar peduncle integrity, the kinetic component of the ICARS was strongly associated with integrity of the superior cerebellar peduncle. Thus, ataxia assessments reflect cerebellar communication with the cortex via the superior cerebellar peduncle, whereas balance assessment reflect cerebellar communication with spinal proprioceptive inputs.

Study limitations Postural sway deficits observed in PwMS could be attributed to the involvement of additional structures other than just the cerebellum. In addition, we cannot rule out whether other neurologic deficits, such as spasticity or muscle weakness also contributed to abnormal postural sway in our cohort.

Conclusions In summary, we show that deficits in the superior cerebellar peduncle contribute to the postural sway deficits observed when standing with visual input available, as well as to clinical measures of ataxia, whereas deficits in the inferior cerebellar peduncle contributes to postural sway deficits with or without visual inputs, as well as to clinical measures of postural responses. Thus, balance rehabilitation should take into consideration that specific locations of MS disruption of white matter tracts result in different types of balance disorders, and likely environmentally-specific falls. In addition, derivatives of postural sway acceleration (ie, jerk) can serve as a sensitive measure of postural control related to integrity of the cerebellar peduncles.

Suppliers a. b. c. d.

Magnetom Tim Trio 3.0T scanner; Siemens Healthineers. FMRIB Software Library v. 5; FMRIB Analysis Group. Opal wearable sensor; APDM Wearable Technologies, Inc. Velcro belt; Velcro.

www.archives-pmr.org

Postural sway deficits multiple sclerosis

7

Keywords Balance; Body-worn sensors; Cerebellum; Diffusion-tensor imaging; Inertial sensors; Multiple sclerosis; Neuroimaging; Postural control; Quiet Standing; Rehabilitation

9.

10.

Corresponding author

11.

Geetanjali Gera, PhD, PT, 204L CHS Building, 900 South Limestone, Lexington, KY 40536-0200. E-mail address: [email protected].

12.

Acknowledgement

13.

We thank our subjects for participation and the funding sources. 14.

References 15. 1. Cattaneo D, De Nuzzo C, Fascia T, Macalli M, Pisoni I, Cardini R. Risks of falls in subjects with multiple sclerosis. Arch Phys Med Rehabil 2002;83:864-7. 2. Finlayson ML, Peterson EW, Cho CC. Risk factors for falling among people aged 45 to 90 years with multiple sclerosis. Arch Phys Med Rehabil 2006;87:1274-9. 3. Prosperini L, Petsas N, Raz E, et al. Balance deficit with opened or closed eyes reveals involvement of different structures of the central nervous system in multiple sclerosis. Mult Scler 2014;20:81-90. 4. Fling BW, Dutta GG, Schlueter H, Cameron MH, Horak FB. Associations between proprioceptive neural pathway structural connectivity and balance in people with multiple sclerosis. Front Hum Neurosci 2014;8:814. 5. Spain RI, St George RJ, Salarian A, et al. Body-worn motion sensors detect balance and gait deficits in people with multiple sclerosis who have normal walking speed. Gait Posture 2012;35:573-8. 6. Mancini M, Salarian A, Carlson-Kuhta P, et al. ISway: a sensitive, valid and reliable measure of postural control. J Neuroeng Rehabil 2012;9:59. 7. Peterka RJ. Chapter 2 - Sensory integration for human balance control. In: Day BL, Lord SR, editors. Handbook of clinical neurology. Vol. 159. New York: Elsevier; 2018. p 27-42. 8. Anderson VM, Wheeler-Kingshott CA, Abdel-Aziz K, et al. A comprehensive assessment of cerebellar damage in multiple sclerosis

www.archives-pmr.org

16.

17.

18. 19.

20.

21.

using diffusion tractography and volumetric analysis. Mult Scler 2011; 17:1079-87. Preziosa P, Rocca MA, Mesaros S, et al. Relationship between damage to the cerebellar peduncles and clinical disability in multiple sclerosis. Radiology 2014;271:822-30. Storey E, Tuck K, Hester R, Hughes A, Churchyard A. Inter-rater reliability of the International Cooperative Ataxia Rating Scale (ICARS). Mov Disord 2004;19:190-2. Franchignoni F, Horak F, Godi M, Nardone A, Giordano A. Using psychometric techniques to improve the Balance Evaluation Systems Test: the mini-BESTest. J Rehabil Med 2010;42: 323-31. Eickhoff SB, Jbabdi S, Caspers S, et al. Anatomical and functional connectivity of cytoarchitectonic areas within the human parietal operculum. J Neurosci 2010;30:6409-21. Basser PJ, Pierpaoli C. Microstructural and physiological features of tissues elucidated by quantitative-diffusion-tensor MRI. J Magn Reson B 1996;111:209-19. Gulani V, Webb AG, Duncan ID, Lauterbur PC. Apparent diffusion tensor measurements in myelin-deficient rat spinal cords. Magn Reson Med 2001;45:191-5. Lowe MJ, Beall EB, Sakaie KE, et al. Resting state sensorimotor functional connectivity in multiple sclerosis inversely correlates with transcallosal motor pathway transverse diffusivity. Hum Brain Mapp 2008;29:818-27. Pijnenburg M, Caeyenberghs K, Janssens L, et al. Microstructural integrity of the superior cerebellar peduncle is associated with an impaired proprioceptive weighting capacity in individuals with nonspecific low back pain. PLoS One 2014;9:e100666. Gera G, Chesnutt J, Mancini M, Horak FB, King LA. Inertial sensorbased assessment of central sensory integration for balance after mild traumatic brain injury. Mil Med 2018;183:327-32. Peterka RJ. Sensorimotor integration in human postural control. J Neurophysiol 2002;88:1097-118. El-Gohary M, Peterson D, Gera G, Horak FB, Huisinga JM. Validity of the instrumented push and release test to quantify postural responses in persons with multiple sclerosis. Arch Phys Med Rehabil 2017;98:1325-31. Cameron MH, Horak FB, Herndon RR, Bourdette D. Imbalance in multiple sclerosis: a result of slowed spinal somatosensory conduction. Somatosens Mot Res 2008;25:113-22. Horak FB, Dickstein R, Peterka RJ. Diabetic neuropathy and surface sway-referencing disrupt somatosensory information for postural stability in stance. Somatosens Mot Res 2002;19:316-26.