Muscle Strength, Balance, and Indoor Mobility in Oculopharyngeal Muscular Dystrophy: An Exploratory Canadian Multicenter Study

ABSTRACT

Introduction/Aims

Oculopharyngeal muscular dystrophy (OPMD) is a neuromuscular disease presenting with dysphagia, ptosis, and proximal weakness. Muscle strength and indoor mobility capacity have rarely been studied using standardized clinical outcome assessments (COAs). The objectives of this study were to (1) document muscle strength, balance, and indoor mobility, and (2) explore factors contributing to indoor mobility limitations.

Methods

This cross‐sectional study included 113 participants (mean age = 61.8 years) from four Canadian neuromuscular disease clinics (Saguenay, Quebec, Montreal, Ottawa). The COAs included upper and lower limb quantified muscle testing, Activities Balance and Confidence Scale, BERG Balance Scale, and for indoor mobility the 10‐m Walk Test at self‐selected (10mWTs‐s) and maximal speed (10mWTm), 30‐s Sit‐to‐Stand test (30s‐STS), and 10‐step stair test. Multiple linear regression models using backward selection were used to explore factors associated with each indoor mobility COA.

Results

Muscle strength in percentage of age‐ and sex‐matched reference values ranged from 41.4% to 81.4%, hip flexors being the most severely impaired muscle group. The 10mWTs‐s, 10mWTm, 30s‐STS and 10‐step stair test had percentages of reference values ranging from 44.9% to 83.3%. Balance, balance confidence, and lower limb muscle strength were found to be significant factors for all indoor mobility COAs. All models explained between 57.4% and 79.9% of the variance.

Discussion

This study provides clinically relevant outcome measures for assessing mobility in OPMD and identifies key factors that should be targeted in rehabilitation. These findings may also inform the selection of outcomes for future interventional studies.

Abbreviations

10mWTm

10‐m walk test maximal speed

10mWTs‐s

10‐m walk test self‐selected speed

30s‐STS

30 s sit‐to‐stand test

ABC‐Scale

Activities Balance and Confidence Scale

BBS

Berg Balance Scale

BMI

Body‐mass Index

CIS‐FSS

Checklist Individual Strength—Fatigue Severity Subscale

DM1

Myotonic Dystrophy type 1

HHD

Hand‐held dynamometer

Mini‐BEST

Mini Balance Evaluation Systems Test

N

Newtons

Nm

Newton‐meters

OPMD

oculopharyngeal muscular dystrophy

PABPN1

Poly(A) Binding Protein Nuclear 1

QMT

quantitative muscle testing

SD

standard deviation

TUG

Timed Up and Go

VIF

variance inflation factor

yo

years old

1. Introduction

Oculopharyngeal muscular dystrophy (OPMD) is an autosomal dominant genetic progressive neuromuscular disease with an age of onset in the fifth decade [1]. Proximal muscle weakness has been described as part of the cardinal signs of the disease, along with dysphagia and ptosis [2, 3].

Based on a recent scoping review [4], only five studies have used quantitative muscle testing (QMT) to assess muscle strength [5, 6, 7, 8, 9] and even fewer its impact on indoor mobility activities [9]. The most extensive retrospective study to date (n = 333) reported global proximal lower limb weakness in 86.7% of patients [2]. Impairments were found in lower limb (hip and knee flexors and extensors) and upper limb (shoulder abductors and elbow flexors) muscle strength (measured using QMT), as well as in grip and pinch strength [5, 9]. Furthermore, a significant decline in strength has also been observed in the hip flexors, knee extensors, and deltoids, ranging from 12.2% to 26.8% over a 20‐month period [8]. Distal lower limb muscle weakness has also been reported using QMT, notably in the ankle plantar flexors and dorsiflexors [5].

Regarding indoor mobility, one study reported activity limitations related to sit‐to‐stand transfer and ascending and descending stairs, while self‐selected walking speed was close to what is expected for people of the same age and sex [9]. In another study, a longer time to perform the Timed Up and Go test was reported for individuals with OPMD compared with nonaffected controls [6].

The same scoping review found no study documenting explanatory factors of indoor mobility, and only three correlational studies have been published to date [9, 10, 11]. Moderate to very strong associations have been found between lower limb muscle strength (hip and knee flexors and extensors) and walking speed and endurance, sit‐to‐stand transfer, and stair climbing and descending [9]. This is particularly important since walking speed has been proposed as the main factor explaining the level of social participation in activities of daily living and social roles [11]. A better understanding of muscle weakness and its impact on indoor mobility will also contribute to clinical trial readiness. The objectives of this study were to (1) document muscle strength, balance, and indoor mobility, and (2) explore factors contributing to indoor mobility limitations in individuals with OPMD.

2. Methods

2.1. Standard Protocol Approvals, Registrations, and Patient Consents

The study was approved by the Ethics Review Board of the Centre intégré universitaire de santé et de services sociaux of Saguenay–Lac‐Saint‐Jean (MP‐25‐2022‐416) and OHRI REB approval (20220044–01). Written informed consent was obtained from all participants.

2.2. Study Design and Participants

This study is part of a larger longitudinal study looking at the natural history of OPMD, and data from the first data collection are included here. Participants from our previous smaller‐scale study [9] were first invited to participate (19 out of 34 accepted). New participants were then recruited from the registry of four Neuromuscular Disease Clinics in Canada (Saguenay, Quebec City, Montreal, and Ottawa), based on a stratified sampling strategy by sex and age groups. The inclusion criteria were: (1) age between 40 and 74 years, (2) a genetic or clinical diagnosis of OPMD, and (3) symptomatic. The age range of inclusion was chosen to capture the early onset of symptoms, which generally appear around 50 years, and was limited to 74 years to prevent excessive fatigue that aging in combination with OPMD could have induced, considering the length of the study protocol. If genetic confirmation of OPMD was not available, participants must have had one of the cardinal symptoms (ptosis, dysphagia, and/or proximal muscle weakness) and a family history of OPMD. Participants with medical conditions that could contribute to motor impairments or health conditions impacting clinical outcome assessment (COA) completion were excluded.

2.3. Evaluation Protocol

Sociodemographic data (age, age at diagnosis, sex, gender, [GCN] repeat size, education, occupation and indoor walking aid) were retrieved from the medical record and a sociodemographic questionnaire. Body mass index (BMI) was calculated from the weight and height taken during the visit. Indoor mobility activities were defined based on the definitions provided by Shumway‐Cook [12], encompassing walking, transfers and ascending and descending stairs. Participants were assessed by trained physiotherapists and physiotherapy technologists using standardized operating procedures for three 3‐h sessions, spaced 2 weeks apart as part of the larger natural history study. The following tests were divided between the three visits, and a standardized order was chosen to prevent fatigue:

Muscle strength: QMT using a portable hand‐held dynamometer (HHD) (Medup, Atlas Médic, Quebec, Canada) was used to assess lower limb (hip flexors, extensors and abductors, and knee flexors and extensors) and upper limb (shoulder abductors and flexors, and elbow flexors) maximal isometric muscle strength during a maximal voluntary isometric contraction. The full HHD protocol has been previously published, and this protocol has demonstrated excellent metrological properties [13]. Muscle strength using a “make test” was collected in Newtons (N) and the lever arm length was measured using standardized anatomical landmarks to give a torque measurement in Newton‐meter (Nm). Two trials per side were performed: if there was a difference of > 10% between the first two trials or if one of the two values was deemed not valid (e.g., contraction performed in the wrong plane of movement or obvious compensation), additional trials were performed until < 10% difference was obtained between two trials up to a maximum of five trials. The mean of the two closest values in Nm for each side was calculated, and the overall mean of both sides was used for final analyses. Handgrip strength (kg) was assessed using a Jamar digital dynamometer (Asimow Engineering Co., Los Angeles, CA), whereas lateral pinch strength (kg), defined as the force generated between the thumb and index finger, was assessed using a digital pinch gauge (Baseline Pinch Gauge, Fabrication Enterprises Inc., Irvington, NY). For each one, three maximal voluntary contractions were recorded for each hand, and the mean was computed.

2.3.1. Balance

Self‐efficacy and balance perception: The Activities Balance Confidence Scale—Simplified Questionnaire (ABC‐Scale) [14] is a patient‐reported outcome measure assessing the participant's confidence not to lose their balance during various tasks of daily living. It includes 15 items scored on a 4‐point Likert scale for a maximum score of 45 points, where a higher score indicates better balance confidence, and a score < 30 indicates a risk of falls.

Balance: The BERG Balance Scale (BBS) [15] is a clinician‐reported outcome measure including 14 items, each scored from 0 to 4, where a higher score indicates better performance, for a maximum of 56. A score < 45 is indicative of a high risk of falls. If the score was higher than the cutoff score of 45, participants also performed the Mini‐BEST test. The Mini Balance Evaluation Systems Tests (Mini‐BEST) [16] is a clinician‐reported outcome measure including 14 items. The total Mini‐BEST maximal score is 28, where a higher score indicates better performance.

Falls: Self‐reported number of falls during the last 12 months were documented. Risk of fall was reported according to cutoff score of the ABC‐Scale (< 30) [17] and BBS (< 45) [18].

Fatigue: The Checklist Individual Strength—Fatigue Severity Subscale (CIS‐FSS) contains eight items, each scored on a seven‐point Likert scale, with a total score ranging from 8 to 56, where a higher score means a higher level of fatigue. A score ≥ 35 was used as cutoff to discriminate severe fatigue [19].

2.3.2. Indoor Mobility Activities

Walking speed: The 10‐m walk test was performed at self‐selected (10mWTs‐s) and maximal (10mWTm) speed. The time to cover a 10‐m distance over a 14‐m corridor was recorded and converted into meters/s for each pace.

Sit‐to‐stand transfer capacity: The 30‐s sit‐to‐stand test (30s‐STS) consists of rising from sitting to standing position as many times as possible within 30 s. Two or three attempts were allowed to obtain two valid trials. The mean number of complete sit‐to‐stand repetitions without arm support was calculated and used for analyses. If a participant was unable to perform a repetition without support, the score is 0.

Stair walking: The 10‐step stair test assesses the time (in seconds) required to ascend and descend (separately) 10 steps at comfortable and maximal speed. The use of handrails was also recorded.

Functional mobility: The TUG test assesses the time required to stand from a chair, walk back and forth a 3‐m distance, and sit back at self‐selected speed. The time in seconds to complete the task was recorded.

2.4. Statistical Analyses

Descriptive statistics for continuous variables (mean and standard deviation [SD]) and categorical variables (frequency and percentage) were documented. Distribution was tested for normality using the Kolmogorov–Smirnov test. Differences between sexes were assessed using the Mann–Whitney U test for nonparametric distributions and Student's t‐test for parametric distributions for all COA using raw data and percentages of reference values (when available). The risk of falls, according to documented cutoff scores obtained from healthy older adults, was reported as number and percentage. To compare each participant's observed performance with what is expected in healthy individuals and to provide an indication of the degree of impairment, the values obtained were compared with age‐ and sex‐specific normative reference data from the literature, and expressed as the percentage of these normative values (raw data/reference value × 100), where 100% corresponds to the expected performance for people of the same age and sex as the participant. For timed tests, the percentage was calculated in reverse so that better performance corresponds to values above 100% (reference value/raw data × 100). Reference values and cutoff scores were selected from the following studies for upper and lower limb muscle strength QMT (Hebert, L. J. et al., unpublished manuscript), ABC‐Scale [17], BBS [18], for > 60 years old; [20], for < 60 years old; [21], CIS‐FSS [19], Mini‐BEST [22], 10mWTs‐s and 10mWTm [23], 30s‐STS [24, 25], TUG [26, 27], 10‐step max climb and 10‐step max descent [28, 29]. Percentages of reference values for muscle strength are available only for participants aged 40–69 years, as there are no reference values in healthy subjects aged ≥ 70 using the same standardized operating procedure. Data entry was done in REDCap 12.4.2 software (Vanderbilt University, Nashville, Tennessee). Missing data were handled using pairwise deletion.

For documenting contributing factors of indoor mobility activities, linear regression models using backward selection were conducted for each indoor mobility COA. Each variable was selected based on studies in the aging population [30, 31, 32] and populations with a neuromuscular disease [33]. Correlations between indoor mobility COA and potential contributing factors were performed with the Pearson correlation coefficient (r). Correlations between independent variables were also performed to check for multicollinearity (r > 0.9) except for two‐joint muscle groups, as they are closely related. Confounding (age, sex, BMI) and moderating variables (CIS‐FSS, ABC‐Scale and BBS) were systematically included in each model to control their impacts, and a description of their associations with the dependent variables is included. Despite being assessed, the Mini‐BEST was not included in the models since not everyone was able to perform it. In addition, given their clinical relevance, all five lower limb muscle strength groups available in the natural history study were included in the models. Independent variables were included in three blocks (1—Confounding, 2—Moderating, 3—Muscle strength), and backward selection was applied to reduce the risk of type II errors from the bottom‐up method [34]. Using this method, only significant factors with their respective β coefficients remained in the filtered model.

The final models were verified for multicollinearity using the variance inflation factor (VIF) (should be approximately < 10). Distribution of residuals was tested for normality using the Kolmogorov–Smirnov test, with a significance threshold set at p < 0.05. In case of violation of normality, weighted least‐squared generalized regression (WLS weight) correction was applied and presented accordingly. If the normality of residuals was still not respected after WLS‐adjusted correction, log10 transformation using a log–level regression was made, where a log10 transformation was made only on the dependent variable. For each variable included in the final models, effect size was reported as Cohen's ƒ², which can be interpreted as follows: Small effect ≥ 0.02; Medium effect ≥ 0.15; Large effect ≥ 0.35 [35]. Statistical analyses were performed using SPSS Statistics for Windows, Version 30.0 (IBM, Armonk, NY).

3. Results

A total of 113 participants were included in the study, and all characteristics of the sample are presented in Table 1. Most of the participants had the (GCN)13 mutation, and 53.1% were women (same for sex as for gender). Forty‐two participants reported having fallen in the last 12 months, with a mean of 3.3 falls. Almost half of the participants reported not using walking aids. Walls and furniture were the most used supports across all reported walking aids, and 20.4% used traditional walking aids.

TABLE 1.

Characteristics of participants (n = 113).

	Total	Men (n = 53)	Women (n = 60)
Age, years
Mean (SD)	61.8 (8.2)	62.9 (7.1)	60.9 (9.0)
[min–max]	[41–74]	[49–74]	[41–74]
Age of diagnosis, years (n = 111)		n = 53	n = 58
Mean (SD)	52.0 (7.7)	53.3 (1.0)	50.8 (1.1)
[min–max]	[34–71]	[40–69]	[34–71]
Sex, n (%)
Men	53 (46.9)
Women	60 (53.1)
(GCN) repeats size, n (%)
(GCN)11(GCN)13	1 (0.9)	1 (1.9)	0
(GCN)10(GCN)12	1 (0.9)	1 (1.9)	0
(GCN)10(GCN)13	97 (85.8)	43 (81.1)	54 (90.0)
(GCN)10(GCN)15	2 (1.8)	2 (3.8)	0
(GCN)10(GCN)16	1 (0.9)	1 (1.9)	0
Unknown a	11 (9.7)	5 (9.4)	6 (10.0)
Education, n (%)
Elementary school	2 (1.8)	0 (0)	2 (3.3)
High school	32 (28.3)	20 (37.7)	12 (20.0)
General and professional teaching college	37 (32.7)	15 (28.3)	22 (36.7)
University	32 (28.3)	16 (30.2)	16 (26.7)
Prefer not to answer	10 (8.8)	2 (3.8)	8 (13.3)
Occupation, n (%) (n = 111)			n = 58
Working	35 (31.5)	14 (26.4)	21 (36.2)
Unemployed (for health reasons)	12 (10.8)	7 (13.2)	5 (8.6)
Retired	54 (48.6)	28 (52.8)	26 (44.8)
Early retirement (for health reasons)	5 (4.5)	3 (5.7)	2 (3.4)
Other	2 (1.8)	0 (0)	2 (3.4)
Prefer not to answer	3 (2.7)	1 (1.9)	2 (3.4)
Indoor walking aid, n (%) (n = 113)
None	55 (48.7)	22 (41.5)	33 (55.0)
Foot orthosis	4 (3.5)	3 (5.7)	1 (1.7)
Leaning on walls and furniture	35 (31.0)	15 (28.3)	20 (33.3)
Cane/crutch	9 (8.0)	6 (11.3)	3 (5.0)
Walker/roller	8 (7.1)	5 (9.4)	3 (5.0)
Manual wheelchair	0 (0)	0 (0)	0 (0)
Motorized wheelchair	2 (1.8)	2 (3.8)	0 (0)
Fall reported in the last year, n (%) (n = 112)			n = 59
No	70 (61.9)	31 (58.5)	39 (66.1)
Yes	42 (37.5)	22 (41.5)	20 (33.9)
Number of falls in the last year (n = 42)
Mean (SD)	3.3 (3.3)	2.9 (2.7)	3.7 (3.9)
Median	2.0	2.0	2.0
[min–max]	[1–16]	[1–12]	[1–16]

Note: Missing data: participants did not answer the question.

Abbreviations: min–max, minimal and maximal value (range); SD, standard deviation.

^a Clinical diagnosis or waiting for PABPN1 test results.

3.1. Muscle Strength

Muscle strength in Nm and percentage of reference values are presented in Tables 2 and 3. Significant differences in Nm were found between sexes for all muscle groups, but none when compared using percentages of reference values. The hip flexors were the most affected, with no participant above the reference values. Upper limbs were less affected than lower limbs.

TABLE 2.

Lower limb muscle strength in Nm (raw data) and percentage of reference values for the total sample and divided by sex. ^a

Muscle group	Total sample	Men	Women	p c
Hip flexors
Raw data, Nm (n = 103)
Mean R/L (SD)	50.6 (28.2)	55.5 (32.6)	40.3 (17.3)	0.012
[min–max]	[2.2–140.0]	[2.2–140,0]	[7.4–78.4]
% ref. values (n = 83)		n = 40	n = 43
Mean (SD)	41.4 (20.0)	40.8 (23.3)	42.0 (16.6)	0.695
[min–max]	[1.6–89.0]	[1.6–89.0]	[7.4–78.4]
# above ref. values (%)	0	0	0
Hip extensors b
Raw data, Nm (n = 106)
Mean R/L (SD)	110.6 (68.8)	135.6 (80.2)	87.3 (46.1)	< 0.0001
[min–max]	[16.0–302.4]	[16.0–302.4]	[20.1–208.1]
% ref. values (n = 85)		n = 41	n = 44
Mean (SD)	61.7 (30.1)	60.8 (33.3)	62.6 (27.2)	0.765
[min–max]	[6.9–130.8]	[6.9–130.8]	[20.7–116.6]
# above ref. values (%)	9 (10.6)	5 (12.2)	4 (9.1)
Hip abductors
Raw data, Nm (n = 105)
Mean R/L (SD)	58.8 (33.7)	71.6 (37.9)	47.6 (24.8)	< 0.0001
[min–max]	[6.0–175.0]	[20.8–175.0]	[6.0–115.2]
% ref. values (n = 84)		n = 40	n = 44
Mean (SD)	61.5 (29.3)	62.3 (32.7)	60.8 (26.1)	0.979
[min–max]	[7.8–134.7]	[17.4–134.7]	[7.8–116.6]
# above ref. values (%)	13 (15.5)	7 (17.5)	6 (13.6)
Knee flexors
Raw data, Nm (n = 107)
Mean R/L (SD)	48.8 (26.8)	59.2 (31.6)	39.7 (17.5)	0.001
[min–max]	[4.0–121.9]	[4.0–121.9]	[8.2–86.5]
% ref. values (n = 85)		n = 40	n = 45
Mean (SD)	68.6 (31.8)	68.0 (35.7)	69.1 (28.4)	0.860
[min–max]	[5.1–141.4]	[5.1–141.4]	[14.9–127.8]
# above ref. values (%)	12 (14.1)	7 (17.5)	5 (11.1)
Knee extensors
Raw data, Nm (n = 107)
Mean R/L (SD)	83.6 (46.8)	105.5 (54.3)	64.8 (28.2)	< 0.0001
[min–max]	[12.2–259.6]	[12.2–259.6]	[18.0–156.4]
% ref. values (n = 85)		n = 40	n = 45
Mean (SD)	76.4 (34.1)	78.9 (35.7)	74.2 (31.0)	0.616
[min–max]	[9.0–161.0]	[9.0–161.0]	[22.0–157.1]
# above ref. values (%)	22 (25.9)	12 (30.0)	10 (22.2)

Note: Reason for missing data: poor neuromuscular control, stabilization impossible, paresis, fatigue, and pain.

Abbreviations: % ref. value, percentage of reference values according to sex and age group; min–max, minimal and maximal value (range); Nm, Newton‐meter; R/L, right and left side; SD, standard deviation.

^a Reference values are only available in decades for 40–49, 50–59 and 60–69 years old. 23 participants were 70 or above and were not reported in reference values.

^b Muscle strength was assessed in a different position than the standardized operating procedure used in the reference values study.

^c Difference between men and women using the Mann–Whitney U test.

TABLE 3.

Upper limb muscle strength in Nm (raw data) and percentage of reference values for the total sample and divided by sex. ^a

	Total sample	Men	Women	p b
Hand grip
Raw data, kg (n = 112)		n = 52	n = 60
Mean (SD)	28.2 (10.8)	34.6 (9.8)	22.6 (8.3)	< 0.001
[min–max]	[7.8–68.7]	[11.8–55.0]	[7.8–68.7]
% ref. values (n = 112)		n = 52	n = 60
Mean (SD)	74.5 (21.2)	73.4 (18.5)	75.4 (23.4)	0.930
[min–max]	[26.8–203.0]	[28.6–105.8]	[26.8–203.0]
# above ref. values (%)	6 (5.4)	4 (7.7)	2 (3.3)
Pinch
Raw data, kg (n = 111)		n = 52	n = 59
Mean (SD)	6.3 (2.5)	7.6 (2.6)	5.1 (1.7)	< 0.001
[min–max]	[2.1–14.3]	[2.6–14.3]	[2.1–9.1]
% ref. values (n = 111)		n = 52	n = 59
Mean (SD)	81.4 (25.5)	83.3 (26.6)	79.7 (24.6)	0.425
[min–max]	[31.4–159.8]	[31.6–159.8]	[31.4–138.7]
# above ref. values (%)	18 (16.2)	7 (13.5)	11 (18.6)
Shoulder abductors
Raw data, Nm (n = 107)		n = 50	n = 57
Mean (SD)	47.6 (31.4)	68.6 (33.7)	29.2 (11.2)	< 0.001
[min–max]	[10.6–149.0]	[12.5–149.0]	[10.6–61.0]
% ref. values (n = 84)		n = 39	n = 45
Mean (SD)	74.6 (30.6)	78.3 (35.4)	71.4 (25.7)	0.487
[min–max]	[12.6–147.0]	[12.6–142.2]	[33.2–147.0]
# above ref. values (%)	20 (23.8)	12 (30.8)	8 (17.8)
Shoulder flexors
Raw data, Nm (n = 108)		n = 52	n = 56	< 0.001
Mean (SD)	43.9 (25.1)	59.2 (26.7)	29.7 (11.9)
[min–max]	[9.2–122.0]	[10.5–122.0]	[9.2–53.0]
% ref. values (n = 86)		n = 41	n = 45
Mean (SD)	77.3 (29.2)	76.8 (30.8)	77.8 (28.2)	0.785
[min–max]	[23.4–162.6]	[23.4–162.6]	[32.1–122.2]
# above ref. values (%)	22 (25.6)	10 (24.4)	12 (26.7)
Elbow flexors
Raw data, Nm (n = 106)		n = 50	n = 56
Mean (SD)	33.6 (20.1)	45.6 (22.6)	22.9 (8.2)	< 0.001
[min–max]	[2.8–146.6]	[2.8–146.6]	[6.4–43.6]
% ref. values (n = 83)		n = 39	n = 44
Mean (SD)	79.4 (30.5)	84.5 (36.6)	74.9 (23.3)	0.192
[min–max]	[13.8–235.2]	[13.8–235.2]	[28.3–124.5]
# above ref. values (%)	16 (19.3)	11 (28.2)	5 (11.4)

Note: Reason for missing data: poor neuromuscular control, stabilization impossible, paresis, fatigue, and pain.

Abbreviations: % ref. value, percentage of reference values according to sex and age group; min–max, minimal and maximal value (range); Nm, Newton‐meter; R/L, right and left side; SD, standard deviation.

^a Reference values are only available in decades for 40–49, 50–59 and 60–69 years old. 23 participants were 70 or above and were not reported in reference values.

^b Difference between men and women using the Mann–Whitney U test.

3.2. Indoor Mobility Activities

Table 4 shows the total score and percentage of reference values (when available) for indoor mobility activity limitations. No significant difference was found using raw data between sexes for all activity limitations COA (see Table S1). The most affected COA was the 30s‐STS, with 32 participants who were unable to perform a single repetition without arm support, and a mean of 44.9% of what is expected for people of the same sex and age. Participants showed slightly more difficulty in ascending 10 steps at maximal speed when compared to descending at the same pace.

TABLE 4.

Raw scores and percentage of reference values for indoor mobility COA.

	Raw scores	% of reference values
10mWTs‐s, m/s (n = 110)
Mean (SD)	1.13 (0.34)	83.3 (24.2)
[min–max]	[0.37–1.95]	[27.5–139.5]
10mWTm, m/s (n = 108)
Mean (SD)	1.51 (0.49)	77.4 (23.3)
[min–max]	[0.41–2.97]	[21.2–143.4]
30s‐STS, repetitions (n = 113)
Mean (SD)	8.2 (6.5)	44.9 (34.3)
[min–max]	[0–25.50]	[0–155.5]
Total incapacity (n, (%))	32 (28.3)
10‐step stair test, s
Comfortable climb (n = 100)
Mean (SD)	10.42 (7.56)	N/A
[min–max]	[4.59–47.67]
Use of handrails, n (%)	53 (53.0)
Comfortable descent (n = 99)
Mean (SD)	8.19 (5.31)	N/A
[min–max]	[3.58–32.75]
Use of handrails, n (%)	48 (48.5)
Maximal climb (n = 100)
Mean (SD)	8.20 (6.54)	58.0 (27.0) a
[min–max]	[2.75–37.02]	[9.1–122.2]
Use of handrails, n (%)	56 (56.0)
Maximal descent (n = 97)
Mean (SD)	6.55 (4.99)	67.6 (28.2) a
[min–max]	[2.50–35.31]	[9.6–135.2]
Use of handrails, n (%)	54 (55.7)
TUG, s (n = 107)
Mean (SD)	10.88 (5.87)	100.2 (29.3) a
[min–max]	[5.27–42.70]	[21.6–169.3]
Risk of fall (> 13.5), n (%)	20 (18.7)

Note: Reasons for missing data: low luminosity in the staircase, incapacity to realize the test, pain, fatigue, and lack of time.

Abbreviations: 10mWT, 10‐m walk test; 30s‐STS, 30‐s sit‐to‐stand test; m/s, meters/s; max, maximum; min, minimum; N/A, Nonapplicable; s, second; SD, standard deviation; TUG, Timed Up and Go.

^a For timed tests, percentages from reference values were calculated in reverse so that better performance is also indicated by a value greater than 100% of the norm.

Table 5 presents additional control and moderating variables of indoor mobility limitations in OPMD, which did not show significant differences between sexes (see Table S2). Severe fatigue, as measured by the CIS‐FSS, was present in 37 participants. Risk of falls was present in 33 and 17 participants according to the ABC‐Scale and BBS, respectively. Also, a significant difference between participants who reported falling in the last year and those who did not report any falls was found for the BSS (mean = 45.3 and 52.8, respectively; p < 0.001), the ABC‐Scale (mean = 29.3 and 37.4, respectively; p < 0.001), and the Mini‐BEST (mean = 20.2 and 24.2, respectively; p < 0.001), the score of participants having a fall history being lower than the other participants for the three scales.

TABLE 5.

Raw score, percentage of reference values and cutoff scores for independent variables of indoor mobility limitations.

	Raw scores	% of reference values
BMI, kg/m2 (n = 110)
Mean (SD)	25.9 (5.3)	N/A
[min–max]	[13.7–43.2]
CIS‐FSS, /56 (n = 110)
Mean (SD)	29.1 (11.4)
[min–max]	[8–54]	N/A
Severe fatigue, n (%)
No	73 (66.4)	—
Yes	37 (33.6)	—
ABC‐Scale, /45 (n = 112)
Mean (SD)	34.38 (9.64)
[min–max]	[8–45]	N/A
Risk of falls (< 30), n (%)
No	79 (70.5)	—
Yes	33 (29.5)	—
BBS, /56 (n = 113)
Mean (SD)	50.0 (10.6)	90.8 (19.1)
[min–max]	[0–56]	[0–105.7]
Risk of falls (< 45), n (%)
No	95 (85.0)	—
Yes	17 (15.0)	—
Mini‐BEST, /28 (n = 89)		n = 75 a
Mean (SD)	22.9 (4.0)	92.1 (15.9)
[min–max]	[9–28]	[42.9–128.6]

Note: Reasons for missing data: lack of time, incapacity to reach assessment position due to pain, and fatigue.

Abbreviations: ABC‐scale, Activities and Balance Confidence Scale; BBS, Berg balance scale; BMI, body mass index; CIS‐FSS, Checklist Individual Strength Fatigue Severity Subscale; max, maximum; min, minimum; N/A, nonapplicable; SD, standard deviation.

^a Reference values only available for 50–89 years old.

3.3. Factors Determining Indoor Mobility Activities

Table 6 shows that all independent (control, moderating and lower limb muscle strength) variables were significantly correlated to indoor mobility activities, except for sex and BMI, which were nonetheless included in the models, as the literature showed that indoor mobility is affected by these two variables across different populations [2, 31, 36, 37]. Regarding lower limb muscle strength, a high correlation was found between all muscle groups (r = 0.783–0.915, p < 0.001; see Table S3). When looking at the associations between lower limb muscle strength and indoor mobility variables, hip extensors had the highest correlations with the 30s‐STS, 10mWTs‐s and 10mWTm. The knee flexors muscle group had the highest correlations with TUG, 10‐step max climb and 10‐step max descent.

TABLE 6.

Relationships between independent and indoor mobility variables.

Independent variables	Indoor mobility variables
	10mWTs‐s	10mWTm	30s. STS	10‐step max climb	10‐step max descent	TUG
Control
Age	−0.505 (< 0.001)	−0.539 (< 0.001)	−0.480 (< 0.001)	0.393 (< 0.001)	0.353 (< 0.001)	0.332 (< 0.001)
Sex	0.091 (0.347)	0.014 (0.882)	−0.092 (0.333)	0.053 (0.600)	0.049 (0.633)	−0.095 (0.329)
BMI	−0.076 (0.434)	−0.096 (0.329)	−0.060 (0.535)	−0.024 (0.812)	−0.056 (0.589)	−0.027 (0.782)
Moderating
CIS‐FSS	−0.350 (< 0.001)	−0.346 (< 0.001)	−0.281 (0.003)	0.180 (0.076)	0.238 (0.020)	0.330 (< 0.001)
ABC‐Scale	0.616 (< 0.001)	0.673 (< 0.001)	0.693 (< 0.001)	−0.669 (< 0.001)	−0.661 (< 0.001)	−0.631 (< 0.001)
BBS	0.648 (< 0.001)	0.676 (< 0.001)	0.551 (< 0.001)	−0.745 (< 0.001)	−0.722 (< 0.001)	−0.809 (< 0.001)
Lower limb QMT
Hip flexors	0.413 (< 0.001)	0.532 (< 0.001)	0.645 (< 0.001)	−0.562 (< 0.001)	−0.526 (< 0.001)	−0.491 (< 0.001)
Hip extensors	0.536 (< 0.001)	0.637 (< 0.001)	0.714 (< 0.001)	−0.565 (< 0.001)	−0.513 (< 0.001)	−0.457 (< 0.001)
Hip abductors	0.441 (< 0.001)	0.541 (< 0.001)	0.678 (< 0.001)	−0.498 (< 0.001)	−0.454 (< 0.001)	−0.416 (< 0.001)
Knee flexors	0.482 (< 0.001)	0.584 (< 0.001)	0.693 (< 0.001)	−0.596 (< 0.001)	−0.543 (< 0.001)	−0.509 (< 0.001)
Knee extensors	0.454 (< 0.001)	0.555 (< 0.001)	0.644 (< 0.001)	−0.552 (< 0.001)	−0.514 (< 0.001)	−0.418 (< 0.001)

Note: Results are presented as Pearson r (p‐value). Bold: highest correlation between muscle strength and dependent variable for inclusion in multiple linear regression models.

Abbreviations: 10mWT, 10‐m walk test; 30s‐STS, 30‐s sit‐to‐stand test; ABC‐Scale, Activities and Balance Confidence Scale; BBS, Berg Balance Scale; CIS‐FSS, Checklist Individual Strength—Fatigue Severity Subscale; Nm, Newton‐meter; QMT, Quantitative muscle testing; TUG, Timed Up and Go.

Table S4 presents the six regression models, which initially included all the confounding (age, sex, BMI) and moderating (CIS‐FSS, ABC‐Scale and BBS) variables, as well as the five lower limb muscle groups as potential predictors. For each model, the variables remaining after backward selection were applied are presented. Balance (BBS), balance confidence (ABC‐Scale) and strength of at least one lower limb muscle group always remained significant contributing factors. Furthermore, these variables were among the highest standardized β coefficients in every model, with small to medium effect size for muscle strength and large effect size for balance in most models.

4. Discussion

This large prospective multicenter study demonstrated that muscle strength is impaired in this OPMD population, with lower limbs being more affected than upper limbs. All indoor mobility activities showed limitations, with the capacity to perform sit‐to‐stand transfers being the most affected. Furthermore, lower limb muscle strength and balance were identified as key determinants of indoor mobility limitations.

Regarding muscle strength, OPMD was theorized as a paradigm of premature aging of muscle tissues [38] and our results suggest that its effects exceed the normal aging process and pattern in both men and women, reinforcing this paradigm. Muscle impairment levels reported here are in line with a previous study regarding hip extensors and knee flexors, with greater lower limb impairments [5]. However, hip flexor strength in our cohort was lower than that in the previous study conducted by our research group [9]. One possible explanation is methodological, since Brisson and colleagues used predictive equations [39] based on a slightly different protocol rather than reference values.

Despite significant impairments in muscle strength, values varied widely with some participants exceeding 100% of reference values (except for the hip flexors), particularly in shoulder flexors and knee extensors. This aligns with prior findings showing no difference in knee extensors in a slightly younger cohort of age‐ and sex‐matched controls [6]. This suggests possible compensatory mechanisms that warrant further investigation.

For indoor mobility activities, our cohort showed lower mean performance compared to reference values in all COAs except the TUG, as previously reported [6]. Although the mean performance was within the normal range, 20 participants were categorized as being at fall risk based on TUG results (mean age = 68 years; range 59–74).

According to a systematic review of predictive outcomes in older adults, walking speed is indicative of numerous outcomes. Based on our results, some participants demonstrated walking speed associated with “risk of death, hospitalization and falls” (< 0.7 m/s, n = 12), “mortality, mobility and activity of daily living disability” (< 0.8 m/s, n = 17) and “limited community ambulator” (0.4–0.8 m/s, n = 17) [40]. An increased risk of early mortality has also been reported with a natural walking speed < 0.82 m/s [41], which is the case for 19 patients in our cohort (mean age = 66 years; range 55–74). In addition, maximal walking speed was more impaired than self‐selected. Some participants showed minimal increases in walking speed, with nearly similar comfortable and maximal speeds, possibly due to lower limb weakness [42]. In myotonic dystrophy type 1 (DM1), a small difference between these speeds has been associated with falls [43]. Furthermore, maximal walking speed was reported as an independent predictor of social participation in OPMD [11]. Regarding the factors explaining the walking speed, balance is the strongest predictor, followed by lower limb muscle strength. Balance has been previously associated with walking speed in older adults [44]. It is not surprising that balance contributes significantly to walking speed and other mobility COAs, as most involve activities like walking, changing basic body positions (including transfers) and using stairs.

Sit‐to‐stand transfer capacity is one of the outcomes that were most severely limited. This result aligns with our previous smaller‐scale study [9]. The identified contributing factors were consistent with those found in other populations [45, 46]. Sit‐to‐stand transfers involve hip and knee muscles [12], and substantial use of knee extensors when compared to other lower limb muscles [47]. In the present study, we identified hip muscles as the most important contributors to performance, over knee muscles, consistent with the proximal‐to‐distal progression of weakness in OPMD.

The ability of people with OPMD to ascend and descend stairs was also impaired, with maximal ascent showing more limitations than maximal descent. Also, 10 participants were unable to perform the test, and handrails were frequently used, with a wide range of times required to perform the task. This is in accordance with two cross‐sectional studies [5, 9]. Knee flexion and extension strength have previously been reported to be associated with the time to ascend and descend stairs in older people [48], and we identified hip muscles as important contributors in our OPMD cohort. Ascending stairs requires an extensor moment during the unipodal stance and, considering the greater impairment of the hip extensors, it is consistent with the disease progression pattern that this muscle group explains the most variation in the time to perform the task.

Regarding balance, only one study reported balance from a semi‐structured interview in the literature, in which 7 OPMD participants out of 25 (28.0%) self‐reported poor balance [49]. In our study, scores from the three balance scales aligned with the self‐reported fall history. Balance seems to be an issue in OPMD, as reflected by frequent use of walls and furniture as indoor mobility aids, highlighting the need for further research to assess prevalence and contributing factors.

Altogether, these findings support the development of rehabilitation interventions aimed at improving muscle strength and balance in this population to improve mobility, like those used for other neuromuscular disorders, including DM1 [50].

4.1. Limitations

One limitation of the present study is the lack of metrological properties for COAs specifically documented in the OPMD population. Another limitation is the lack of assessment of some lower limb muscle groups that may affect indoor activities. As this study was a secondary analysis of a natural history study focused on the main impairments in OPMD, which primarily affects proximal muscle groups, key distal muscle groups involved in indoor mobility activities, such as ankle plantar flexors and dorsiflexors, are lacking and should be addressed in future studies. In addition, lower limb coordination was not assessed in the present study, although it can change over the years and increase the risk of falls in older adults [51]. We recommend exploring this concept in future OPMD studies. Another limitation was the relatively long assessment sessions, which could have caused participant fatigue and exhaustion. Finally, the reference values selected may also impact the interpretation of the results, as one participant was compared to a value obtained from healthy persons grouped by decade (e.g., 40–50), and these groups may, in some cases, include a limited number of persons.

5. Conclusion

This study showed strong associations between lower limb muscle strength, balance, balance confidence, and indoor mobility in OPMD. This highlights the need to develop specific rehabilitation interventions aimed at improving muscle strength and balance in this population. Further studies will be needed to better understand all the underlying mechanisms of mobility in this population. Finally, more research is needed to obtain a comprehensive understanding of balance in OPMD.

Author Contributions

Brenda Dennison: investigation. Bernard Brais: conceptualization, writing – review and editing, funding acquisition. Jean‐Denis Brisson: conceptualization, writing – review and editing, funding acquisition. Jessica MacGregor: investigation. Nicolas Bélair: writing – original draft, formal analysis, investigation. Isabelle Lessard: writing – review and editing, conceptualization. Isabelle Côté: writing – review and editing, formal analysis. Elise Duchesne: conceptualization, funding acquisition, writing – review and editing, supervision. Xavier Rodrigue: conceptualization, funding acquisition, writing – review and editing. Cynthia Gagnon: conceptualization, funding acquisition, writing – review and editing, project administration, supervision, data curation. Luc J. Hébert: conceptualization, funding acquisition, writing – review and editing. Hanns Lochmüller: conceptualization, funding acquisition, writing – review and editing.

Funding

This study was funded by the Canadian Institutes of Health Research (#2021PJT‐463584‐MOV‐CFDA‐113059). C.G. and E.D.: Career−grant funding from Fonds de recherche du Québec−santé (no 31011 and no 311186). N.B.: REMDUS excellence master bursary; OPPQ research master bursary, PSDMU research master bursary. H.L. receives support from the Canadian Institutes of Health Research for Foundation Grant FDN‐167281 (Precision Health for Neuromuscular Diseases), Transnational Team Grant ERT‐174211 (ProDGNE) and Network Grant OR2‐189333 (NMD4C), from the Canada Foundation for Innovation (CFI‐JELF 38412), the Canada Research Chairs program (Canada Research Chair in Neuromuscular Genomics and Health, 950‐232279), the European Commission (Grant # 101080249) and the Canada Research Coordinating Committee New Frontiers in Research Fund (NFRFG‐2022‐00033) for SIMPATHIC, and from the Government of Canada, Canada First Research Excellence Fund (CFREF) for the Brain‐Heart Interconnectome (CFREF‐2022‐00007).

Ethics Statement

We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

Conflicts of Interest

C.G. has received consultancy honoraria from Biogen Idec, and acts as a consultant for Seelos, Dyne, and Lupin. She also conducts academic work for Biogen Idec, Ionis, Seelos, and Vertex, all unrelated to the present manuscript. E.D. has consulted or currently consults for Dyne Therapeutics, PepGen, Avidity, AMO Pharma and Vertex Pharmaceuticals, and has received research funding from Vertex Pharmaceuticals, all unrelated to the present manuscript. All other authors have no conflict of interest to declare.

Supporting information

Table S1: Comparison of indoor mobility limitations between sexes.

Table S2: Comparison of control and moderative variables of indoor mobility limitations between sexes.

Table S3: Correlations between strength lower limb muscle groups.

Table S4: Multiple linear regression models of indoor mobility activities.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.