Spatiotemporal gait analysis of patients with spinocerebellar ataxia types 3 and 10 using Inertial Measurement Units: a comparative study

doi:10.21203/rs.3.rs-4266315/v1

Given the high morbidity related to the progression of gait deficits in spinocerebellar ataxias (SCA), there is a growing interest in identifying biomarkers that can guide early diagnosis and rehabilitation. Spatiotemporal parameter (STP) gait analysis using inertial measurement units (IMUs) has been increasingly studied in this context. This study evaluated STP profiles in SCA types 3 and 10, compared them to controls, and correlated them with clinical scales. IMU portable sensors were used to measure STPs under four gait conditions: self-selected pace (SSP), fast pace (FP), fast pace checking-boxes (FPCB), and fast pace with serial seven subtractions (FPS7). Compared to healthy subjects, both SCA groups had higher values for step time, variability, and swing time, with lower values for gait speed, cadence, and step length. We also found a reduction in speed gain capacity in both SCA groups compared to controls and an increase in speed dual-task cost in the SCA10 group. However, there were no significant differences between the SCA groups. Swing time, mean speed, and step length were correlated with disease severity, risk of falling and functionality in both clinical groups. In the SCA3 group, fear of falling was correlated with cadence. In the SCA10 group, results of the Montreal cognitive assessment test were correlated with step time, mean speed, and step length. These results show that individuals with SCA3 and SCA10 present a highly variable, short-stepped, slow gait pattern compared to healthy subjects, and their gait quality worsened with a fast pace, mainly with dual task involvement.

Spinocerebellar ataxias

Gait analysis

Accelerometry

Inertial measurement units

Spatiotemporal gait parameters

Spinocerebellar ataxia (SCA) refers to a group of autosomal dominant, neurodegenerative diseases characterized by the involvement of the cerebellum and its pathways. Approximately 50 types of SCAs have been described, each with its own neurological signs and symptoms and unique causative genetic mutations [1, 2].

SCA3 is the most prevalent type both in Brazil (59.6%) and globally [3]. Expansions in glutamine (CAG) repeats in the Ataxin-3 protein gene (ATXN3) located in the 14q24.3-q32 allele [4, 5] have been linked to five disease phenotypes. These vary according to the signs associated with the most prominent cerebellar syndrome, including dystonia, parkinsonism, spasticity and hyperreflexia, spastic paraparesis, ophthalmoparesis, and peripheral neuropathy [6]. The age of onset is typically 40 years, while the life span of sufferers is approximately 20 years, with a mean progression estimated to reach 1.61 ± 0.12 points/year based on the scale for the assessment and rating of ataxia (SARA) [7].

SCA10 is caused by repeated expansions of the ATTCT pentanucleotide located in intron 9 of the ATXN10 gene on chromosome 22q13.3 [8, 9]. The disease is highly prevalent in Mexico, Peru, and Brazil, with subtle differences in the phenotypes among these populations [10]. In Mexico and Peru, the most common clinical presentation is a cerebellar syndrome associated with epilepsy [11–14]. In Brazil, the most common phenotype is a pure cerebellar syndrome with slow progression (0.44 SARA points/year) [13, 15].

Gait ataxia is the most frequently presenting cerebellar symptom reported by patients and is invariably associated with a progressive loss of functionality [16]. Clinically, gait ataxia is characterized by a wide base, trunk instability, irregular and staggering steps, and an increased risk of falls. Unfortunately, physical examinations and the measurement scales used in clinical practice, such as the SARA and the International Cooperative Ataxia Rating Scale, are limited in their ability to provide accurate qualitative and quantitative data to describe ataxic gait patterns in further detail [17].

Inertial measurement units (IMU) are part of an emerging digital technology arsenal for assessing mobility, including spatiotemporal parameters (STPs) in neurodegenerative diseases. While some studies have evaluated patients with cerebellar ataxia using wearables [18], STPs have not been formally studied in SCA10 patients using IMUs. Furthermore, most studies that have evaluated STPs in ataxic gait patients included etiologically heterogeneous cerebellar disorders as a single group and compared them with controls. Considering that SCA10 patients typically present a more benign overall progression of cerebellar signs and symptoms than other SCAs, we hypothesized that SCA10 would exhibit a comparatively unique spatiotemporal profile. Therefore, this study aimed to analyze the spatiotemporal gait parameters of SCA10 and SCA3 patients, compare these parameters with healthy subjects, and correlate the results with functional clinical scales.

2.1 Patients

Thirty-one patients were included in the study, 16 with SCA10 and 15 with SCA3. One participant in the SCA3 group did not perform the dual-task tests. A control group comprised 24 volunteers aged 18 to 60 without neurological or orthopedic disorders.

Inclusion criteria were as follows: (1) confirmed SCA10 or SCA3 diagnosis established by genetic testing of the patient or a first or second-degree relative; (2) aged between 18 and 60 years; and (3) able to perform a gait test battery without assistance. Exclusion criteria were: (1) severe visual acuity loss; (2) a history of alcohol or substance abuse other than tobacco; (3) a diagnosis of a previous or current orthopedic condition; (4) a diagnosis of another prior or current neurological condition; (5) a history of neurosurgery; (6) current pregnancy; or (7) having a severe cognitive deficit clinically characterized by a Montreal Cognitive Assessment (MoCA) ≤ 5 points.

The same trained examiner collected data during a single study participant visit in the ataxia outpatient unit at the Hospital de Clínicas, Federal University of Paraná (HC-UFPR) between December 2018 to July 2020. The Neurology Department from Kiel University provided the IMUs, training, and the gait test protocol, which was adapted from the ComOn study [19]. All participants signed a voluntary, informed consent form and the study was approved by HC-UFPR’s Committee for Ethics in Human Research.

2.2 Clinical and Gait Assessments

Spatiotemporal gait variables were obtained using three RehaGait^® sensors (Hasomed GmbH, Magdeburg, Germany), two were fastened by Velcro to the feet (one on each foot) and one in the area of the lumbar spine at the L4-L5 level. Each sensor had a triaxial accelerometer (range ± 16 g), a triaxial gyroscope (range ± 2000 °/s), and a magnetometer (range ± 1.6 G). Sensors recorded data at a frequency of 100 times per second and transmitted it in real-time to a tablet via Bluetooth. Each sensor weighed 34 g and was 60x25x15 mm in size. The equipment and algorithm used for the analysis of gait parameters have been previously validated [20–22]. Likewise, the process used to collect the raw data was also a previously validated algorithm [19–20, 23–24].

The gait assessment protocol has been validated in previous studies [19, 25–28] and entails phases of straight walking for 20 meters during single and dual tasking. Single-task performance was assessed during two walking conditions: a self-selected pace (SSP) and a fast pace (FP), in which participants were asked to walk as fast as possible without running. Dual-task performance was assessed under two different conditions: fast pace checking-boxes (FPCB) and fast pace with serial seven subtractions (FPS7). All participants walked while wearing comfortable flat shoes [29].

The following gait parameters were extracted: stride time; step time; step time variability; step time asymmetry; stance time; double support time; double support time variability; swing time; mean speed; cadence; and step length. Step time variability and double support time variability were calculated using a combined standard deviation. Step time asymmetry was defined as the absolute value obtained by subtracting the arithmetic averages of the step time of the left and right feet.

A dual-task cost (DTC) was calculated for each group using the following formula [30]:

$$\text{DTC }\left(\text{%}\right)\text{= ± 100 x }\left[\frac{\left(\text{FP - FP dual task}\right)}{\text{FP}}\right]$$

Percent DTC was calculated for gait speed under the FPCB (GS-FPCB-DTC) and FPS7 (GS-FPS7-DTC) conditions. The ratio between SSP and FP (GS ratio SSP/FP) was also calculated [31].

The following clinical scales were also used in this study: SARA, the Berg balance scale (BBS), the Barthel index (BI), the falls efficacy scale-international (FES-I), and the Montreal cognitive assessment (MoCA). All scales have been previously validated in the Brazilian population [32–36].

2.3 Statistics

Descriptive data are presented as means, standard deviations, and minimum-maximum values. A one-way ANOVA and the Kruskal-Wallis test were used to compare quantitative variables between the three groups. A post hoc analysis was performed using the Student's t-test for independent samples and the nonparametric Mann-Whitney test was used for non-normally distributed data. Spearman's rank correlation coefficient was used to assess the association between quantitative variables. The association between dichotomous/ordinal variables was assessed using the Chi-squared and Fisher's exact tests. A p-value of 0.05 was considered statistically significant. Excel v16.54 and SPSS v21 (IBM Corp., Armonk, NY, USA) were used for all statistical analyses.

3.1 Descriptive Analyses

Gender, age, body mass, height, body mass index, handedness, and MoCA scores did not significantly differ between the three groups. Disease duration, age at onset of motor symptoms, SARA, BBS, BI, and FES-I scales did not significantly differ between the SCA3 and SCA10 groups. Detailed descriptive data is presented in Table 1.

Table 1

Comparison of descriptive data and clinical scales between groups. (*)p < 0,05; (^a) BMI (Body Mass Index); (^b) yos (years of study); (^c) y (years); (^d) yo (years-old); (^e) SARA (Scale for the Assessment and Rating of Ataxia); (^f) BBS (Berg Balance Scale); (^g) BI (Barthel Index); (^h) FES-I (Falls Efficacy Scale – International); (ⁱ) MoCA (Montreal Cognitive Assessment)
Data	SCA10	SCA3	Controls	p-value*
Participants (N)	16	15	24	-
Sex ([%] female) (N)	69 (11)	33 (5)	50 (12)	0,142
Age [years]	45 ± 8,4 (33,2–59,4)	44,8 ± 8,2 (27,3–55,4)	41,7 ± 11,3 (25,3–59,5)	0,482
Genetic testing [%] (N)	69 (11)	47% (7)	-	-
Repeat Expansion size	2088 ± 339 (1573–2680)	70 ± 8 (78 − 52)	-	-
Height [cm]	162 ± 9 (150–186)	166 ± 7 (150–176)	169 ± 11 (145–186)	0,136
Weight [kg]	74,6 ± 16,5 (57–132)	70,1 ± 8,4 (50–84)	77,3 ± 12,7 (47–101)	0,253
BMI [kg/cm2]^a	28,1 ± 3,6 (22,8–38,2)	25,3 ± 2,7 (19,5–29,4)	27,2 ± 3,8 (18,3–35,8)	0,086
Education [yos]^b	11,7 ± 4 (5–18)	10 ± 4,4 (5–21)	14,5 ± 6,3 (4–30)	0,033
Right hand Dominance [%] (n)	75% (12)	93% (14)	79% (19)	0,33
Disease Duration [y]^c	9 ± 6 (1–24)	10 ± 5 (3–24)	-	0,636
Motor Symptoms onset [yo]^d	36 ± 6 (26–47)	35 ± 8 (19–46)	-	0,685
SARA^e	9,3 ± 3,6 (3–18)	10,3 ± 4,8 (3–20)	Not Rated	0,607
BBS^f	48 ± 7,4 (35–56)	43 ± 8,8 (29–56)	Not Rated	0,16
BI^g	96 ± 4 (90–100)	94 ± 4,8 (85–100)	Not Rated	0,446
FES-I^h	31,5 ± 8,2 (22–56)	35,1 ± 8,8 (23–53)	Not Rated	0,147
MoCAⁱ	22,8 ± 4,4 (15–30)	21,9 ± 4,6 (13–27)	25 ± 4,9 (15–30)	0,087

(*)p<0,05; (^a) BMI (Body Mass Index); (^b) yos (years of study); (^c) y (years); (^d) yo (years-old); (^e) SARA (Scale for the Assessment and Rating of Ataxia); (^f) BBS (Berg Balance Scale); (^g) BI (Barthel Index); (^h) FES-I (Falls Efficacy Scale – International); (ⁱ) MoCA (Montreal Cognitive Assessment)

Table 2: Spatiotemporal gait parameters in SCA3, SCA10 and control groups						Post-hoc analysis p-value*
Spatiotemporal Parameter (STP)	Gait test	SCA10 (n = 16)	SCA3 (n = 15)	Control (n = 24)	ANOVA p-values**	SCA10 vs. Control	SCA3 vs. Control	SCA10 vs. SCA3
Stride Time [s]	SSP^a	1,147 ± 0,132 (0,983-1,555)	1,117 ± 0,158 (0,927-1,582)	1,079 ± 0,059 (0,923-1,210)	0,031	0,072	0,013	0,481
	FP^b	1,072 ± 0,159 (0,875-1,560)	1,059 ± 0,131 (0,726-1,290)	0,927 ± 0,049 (0,832-1,045)	<0,001	< 0,001	0,001	0,75
	FPCB^c	1,159 ± 0,177 (0,964-1,619)	1,165 ± 0,147 (0,948-1,482)	0,969 ± 0,055 (0,842-1,097)	< 0,001	< 0,001	< 0,001	0,899
	FPS7^d	1,143 ± 0,126 (0,961-1,452)	1,167 ± 0,137 (1,001–1,498)	0,99 ± 0,053 (0,889-1,063)	< 0,001	< 0,001	< 0,001	0,531
Step Time [s]	SSP^a	0,575 ± 0,066 (0,494-0,779)	0,59 ± 0,081 (0,463-0,805)	0,54 ± 0,03 (0,462-0,605)	0,03	0,068	0,013	0,492
	FP^b	0,537 ± 0,080 (0,443-0,782)	0,529 ± 0,065 (0,364-0,644)	0,464 ± 0,025 (0,416-0,522)	<0,001	< 0,001	0,001	0,721
	FPCB^c	0,581 ± 0,089 (0,483-0,813)	0,582 ± 0,073 (0,477-0,74)	0,485 ± 0,027 (0,42 − 0,548)	<0,001	< 0,001	< 0,001	0,94
	FPS7^d	0,574 ± 0,062 (0,485-0,728)	0,585 ± 0,069 (0,499-0,755)	0,496 ± 0,027 (0,444-0,533)	< 0,001	< 0,001	< 0,001	0,584
Step Time Variability [s]	SSP^a	0,077 ± 0,048 (0,034 − 0,218)	0,081 ± 0,034 (0,038 − 0,176)	0,046 ± 0,014 (0,019 − 0,079)	0,002	0,005	0,002	0,732
	FP^b	0,073 ± 0,038 (0,032 − 0,189)	0,069 ± 0,021 (0,047 − 0,135)	0,048 ± 0,022 (0,023 − 0,104)	0,012	0,007	0,024	0,681
	FPCB^c	0,074 ± 0,032 (0,031 − 0,158)	0,08 ± 0,029 (0,031 − 0,143)	0,047 ± 0,021 (0,019 − 0,096)	0,001	0,003	0,001	0,532
	FPS7^d	0,082 ± 0,069 (0,037 − 0,328)	0,085 ± 0,029 (0,048 − 0,14)	0,048 ± 0,021 (0,023 − 0,106)	0,012	0,016	0,01	0,805
Step Time Asymmetry [s]	SSP^a	0,039 ± 0,024 (0,005 − 0,087)	0,027 ± 0,030 (0,001 − 0,113)	0,022 ± 0,016 (0,002 − 0,059)	0,082	-	-	-
	FP^b	0,022 ± 0,018 (0,003 − 0,061)	0,035 ± 0,035 (0,001 − 0,118)	0,018 ± 0,015 (0,0–0,045)	0,10	-	-	-
	FPCB^c	0,039 ± 0,035 (0,003 − 0,107)	0,048 ± 0,03 (0,002 − 0,102)	0,022 ± 0,017 (0,003 − 0,071)	0,014	0,055	0,005	0,353
	FPS7^d	0,025 ± 0,018 (0,001 − 0,058)	0,033 ± 0,025 (0,006 − 0,09)	0,019 ± 0,017 (0,001 − 0,064)	0,124	-	-	-
Stance Time [s]	SSP^a	1,003 ± 0,115 (0,858-1,349)	1,021 ± 0,127 (0,838-1,351)	0,950 ± 0,055 (0,818-1,066)	0,065	-	-	-
	FP^b	0,933 ± 0,134 (0,774-1,342)	0,923 ± 0,101 (0,676-1,090)	0,818 ± 0,043 (0,736-0,927)	<0,001	< 0,001	0,001	0,781
	FPCB^c	1,009 ± 0,15 (0,842-1,399)	1,015 ± 0,113 (0,869-1,257)	0,855 ± 0,047 (0,748-0,960)	< 0,001	< 0,001	< 0,001	0,88
	FPS7^d	0,997 ± 0,109 (0,826-1,251)	1,012 ± 0,112 (0,862-1,270)	0,870 ± 0,051 (0,771-0,946)	< 0,001	< 0,001	< 0,001	0,643
Double Support Time [s]	SSP^a	0,427 ± 0,049 (0,367-0,569)	0,433 ± 0,052 (0,375-0,575)	0,409 ± 0,028 (0,342-0,462)	0,17	-	-	-
	FP^b	0,397 ± 0,056 (0,33 − 0,563)	0,395 ± 0,039 (0,315-0,45)	0,355 ± 0,02 (0,319-0,401)	0,001	0,002	0,003	0,88
	FPCB^c	0,430 ± 0,063 (0,36 − 0,587)	0,431 ± 0,038 (0,37 − 0,501)	0,369 ± 0,021 (0,321-0,411)	< 0,001	< 0,001	< 0,001	0,949
	FPS7^d	0,427 ± 0,045 (0,348-0,525)	0,429 ± 0,044 (0,362-0,519)	0,374 ± 0,026 (0,314-0,414)	<0,001	< 0,001	< 0,001	0,891
Double Support Time Variability [s]	SSP^a	0,066 ± 0,046 (0,029 − 0,216)	0,066 ± 0,035 (0,030 − 0,160)	0,039 ± 0,012 (0,015 − 0,059)	0,011	0,01	0,013	0,964
	FP^b	0,064 ± 0,037 (0,028 − 0,175)	0,06 ± 0,021 (0,041 − 0,118)	0,042 ± 0,022 (0,018 − 0,097)	0,029	0,015	0,048	0,691
	FPCB^c	0,061 ± 0,028 (0,02 − 0,137)	0,063 ± 0,021 (0,024 − 0,111)	0,042 ± 0,021 (0,015 − 0,091)	0,013	0,015	0,012	0,86
	FPS7^d	0,071 ± 0,067 (0,032 − 0,314)	0,071 ± 0,022 (0,041 − 0,116)	0,042 ± 0,021 (0,018 − 0,102)	0,040	0,031	0,039	0,99
Swing Time [s]	SSP^a	0,148 ± 0,020 (0,126-0,210)	0,156 ± 0,033 (0,088 − 0,231)	0,131 ± 0,012 (0,103-0,159)	0,003	0,019	0,001	0,317
	FP^b	0,140 ± 0,025 (0,110-0,219)	0,135 ± 0,033 (0,049 − 0,194)	0,109 ± 0,01 (0,095 − 0,128)	<0,001	< 0,001	0,001	0,524
	FPCB^c	0,151 ± 0,028 (0,123-0,225)	0,151 ± 0,041 (0,06 − 0,239)	0,115 ± 0,01 (0,099 − 0,139)	< 0,001	< 0,001	< 0,001	0,936
	FPS7^d	0,147 ± 0,02 (0,127-0,203)	0,156 ± 0,03 (0,132-0,236)	0,121 ± 0,01 (0,101-0,148)	<0,001	< 0,001	< 0,001	0,244
Mean Speed [m/s]	SSP^a	0,802 ± 0,234 (0,294-1,107)	0,874 ± 0,233 (0,301-1,23)	1,09 ± 0,135 (0,880-1,556)	< 0,001	< 0,001	0,001	0,313
	FP^b	0,933 ± 0,294 (0,308-1,221)	1,009 ± 0,248 (0,444-1,448)	1,421 ± 0,154 (1,201-1,686)	< 0,001	< 0,001	< 0,001	0,357
	FPCB^c	0,763 ± 0,285 (0,224-1,146)	0,899 ± 0,277 (0,447-1,299)	1,337 ± 0,147 (1,058 − 1,564)	<0,001	< 0,001	< 0,001	0,114
	FPS7^d	0,821 ± 0,268 (0,229-1,071)	0,887 ± 0,197 (0,588-1,152)	1,283 ± 0,169 (0,992-1,575)	<0,001	< 0,001	< 0,001	0,392
Cadence [step/min]	SSP^a	97,42 ± 9,21 (73,19–110,16)	95,57 ± 13,45 (67,99–126,47)	99,95 ± 5,58 (89,69–112,06)	0,352	-	-	-
	FP^b	104,5 ± 11,54 (73,89–120,88)	106,07 ± 17,14 (84,92–160,64)	114,52 ± 6,85 (100,88–131,09)	0,019	0,011	0,034	0,711
	FPCB^c	97,79 ± 11,31 (68,89–113,47)	95,56 ± 11,36 (73,27–116,88)	110,24 ± 6,78 (98,1-126,66)	<0,001	< 0,001	< 0,001	0,527
	FPS7^d	97,89 ± 9,57 (78,45–111,41)	95,47 ± 11,11 (70,87–112,37)	107,32 ± 6,44 (95,24–118,11)	<0,001	0,002	< 0,001	0,455
Step Length [m]	SSP^a	0,49 ± 0,127 (0,24 − 0,69)	0,56 ± 0,147 (0,14 − 0,71)	0,65 ± 0,07 (0,54 − 0,83)	< 0,001	< 0,001	0,071	0,217
	FP^b	0,53 ± 0,146 (0,25 − 0,69)	0,58 ± 0,144 (0,26 − 0,74)	0,75 ± 0,074 (0,63 − 0,91)	< 0,001	< 0,001	0,001	0,988
	FPCB^c	0,46 ± 0,150 (0,18 − 0,67)	0,56 ± 0,138 (0,32 − 0,74)	0,73 ± 0,075 (0,61 − 0,87)	< 0,001	< 0,001	0,002	0,403
	FPS7^d	0,50 ± 0,154 (0,15 − 0,67)	0,56 ± 0,117 (0,34 − 0,77)	0,72 ± 0,083 (0,57 − 0,87)	< 0,001	< 0,001	0,001	0,99
(^)Kruskal–Wallis one-way analysis of variance (p < 0,05); (^*)One-way analysis of variance (p < 0,05); (^a) Self-selected pace; (^b) Fast pace; (^c) Fast Pace checking-boxes; (^d) Fast pace with serial seven subtractions.

3.2 Comparative Analyses

None of the walking paradigms showed significant differences in IMU-based gait parameters between the SCA10 and SCA3 groups. However, compared to controls, both clinical groups exhibited significant differences among specific STPs. Step and stride times were longer in almost all walking tasks except for SSP in the SCA10 group. Longer swing times and higher temporal variability were observed for all tested gait conditions. Mean speed comparisons showed that both SCA groups were slower than the control group in all four tasks. Stance and double support times were significantly longer in both SCA groups except under the SSP condition. Similarly, cadence was significantly lower except under the SSP condition. For all gait conditions, the step length was shorter, except in the SCA3 group under the SSP condition. Moreover, only the SCA3 group under the FPCB condition exhibited increased step time asymmetry compared with controls. There was no difference between the groups for cadence, stance time, or double support time under the SSP condition. The GS-FPCB-DTC was increased only in the SCA10 group when compared to controls. No significant differences were found for the GS-FPS7-DTC. The GS ratio SSP/FP in both SCA groups was higher than in the control group. Detailed results can be found in tables 2 and 3.

Table 3: Dual-task Cost and Single-tasks Ratio					Post-hoc analysis p-values*
Gait Índex	SCA10 (n=16)	SCA3 (n=15)	Control (n=24)	ANOVA p-values**	SCA10 vs. Control	SCA3 vs. Control	SCA10 vs. SCA3
GS-FPCB-DTC (%)^a	19,91 ± 10,836 (6,14-47,22)	15,68 ± 14,114 (-1,81-49,09)	5,72 ± 7,215 (-7,20-20,03)	<0,001	<0,001	0,071	0,604
GS-FPS7-DTC (%)^b	12,28 ± 13,253 (-6,14-54,56)	15,40 ± 10,892 (-3,60-40,55)	9,52 ± 9,193 (-7,08-24,41)	0,261	-	-	-
GS Ratio SSP/FP^c	0,88 ± 0,098 (0,66-1,02)	0,86 ± 0,081 (0,68-1,03)	0,77 ± 0,097 (0,62-1,12)	0,002	0,001	0,006	0,659

(^*)Kruskal–Wallis one-way analysis of variance (p<0,05); (^**)One-way analysis of variance (p<0,05); (^a) Gait speed dual-task cost under the FPCB; (^b) Gait speed dual-task cost under FPS7; (^c) Gait speed ratio between SSP and FP single-tasks.

3.3 Correlations

Table 4 shows correlations between gait variables and the clinical scales. Strong negative correlations were found between the SCA groups, SARA scores and gait speed in all of the tests. With the exception of SCA3 under the FPS7 condition, step length was strongly negatively correlated with SARA scores in both clinical groups. When swing time in FP condition was evaluated with and without dual tasks, both SCA groups positively correlated with SARA. A direct correlation was also found between SARA and step time asymmetry in the SCA10 group under the SSP condition.

Table 4

Correlations between clinical scales and Spatiotemporal Parameters
Spatiotemporal Parameter	Gait Test	SARA^e				BBS^f				BI^g				FES-I^h				MoCAⁱ
		SCA3		SCA 10		SCA 3		SCA 10		SCA 3		SCA 10		SCA 3		SCA 10		SCA 3		SCA 10
		r	p	r	p	r	p	r	p	r	p	r	p	r	p	r	p	r	p	r	p
Stride Time [s]	SSP^a	0,030	0,919	0,370	0,163	-0,020	0,944	-0,450	0,079	-0,010	0,984	-0,140	0,595	0,430	0,110	0,250	0,360	0,130	0,651	-0,630	0,009
	FP^b	0,460	0,085	0,520	0,038	-0,460	0,086	-0,570	0,022	-0,510	0,052	-0,420	0,106	0,720	0,002	0,240	0,381	-0,150	0,587	-0,670	0,005
	FPCB^c	0,470	0,089	0,520	0,037	-0,500	0,067	-0,460	0,072	-0,390	0,172	-0,440	0,092	0,470	0,087	0,140	0,612	0,040	0,893	-0,670	0,005
	FPS7^d	0,510	0,060	0,350	0,180	-0,500	0,067	-0,400	0,124	-0,260	0,365	-0,370	0,159	0,630	0,016	0,120	0,655	-0,190	0,511	-0,330	0,216
Step Time [s]	SSP^a	-0,020	0,955	0,320	0,229	-0,010	0,980	-0,480	0,058	-0,010	0,968	-0,160	0,555	0,330	0,234	0,180	0,507	0,180	0,531	-0,590	0,016
	FP^b	0,400	0,139	0,500	0,050	-0,440	0,102	-0,550	0,026	-0,540	0,036	-0,470	0,064	0,680	0,005	0,190	0,476	-0,110	0,684	-0,660	0,006
	FPCB^c	0,470	0,089	0,500	0,049	-0,500	0,067	-0,440	0,085	-0,390	0,172	-0,450	0,077	0,470	0,087	0,120	0,671	0,040	0,893	-0,660	0,005
	FPS7^d	0,420	0,137	0,310	0,238	-0,480	0,082	-0,420	0,107	-0,300	0,291	-0,390	0,137	0,660	0,010	0,090	0,744	-0,200	0,491	-0,260	0,333
Step Time Variability [s]	SSP^a	0,280	0,314	0,090	0,727	-0,250	0,373	-0,430	0,095	-0,270	0,331	0,160	0,563	0,010	0,960	0,020	0,939	0,210	0,445	0,180	0,511
	FP^b	0,210	0,454	0,420	0,104	-0,110	0,688	-0,440	0,084	-0,310	0,254	-0,280	0,302	0,080	0,780	0,270	0,316	-0,220	0,437	-0,290	0,273
	FPCB^c	0,330	0,249	0,380	0,142	-0,470	0,094	-0,300	0,256	-0,450	0,108	-0,530	0,034	0,350	0,217	-0,020	0,952	-0,030	0,922	-0,160	0,547
	FPS7^d	0,230	0,431	0,190	0,481	-0,450	0,109	-0,320	0,224	-0,610	0,022	-0,340	0,200	0,250	0,391	0,180	0,507	-0,450	0,110	-0,100	0,703
Step Time Asymmetry [s]	SSP^a	-0,140	0,607	0,580	0,019	0,070	0,794	-0,440	0,084	-0,020	0,937	0,070	0,791	-0,350	0,198	0,240	0,372	0,080	0,770	-0,290	0,268
	FP^b	0,200	0,483	0,130	0,626	-0,370	0,172	0,260	0,322	-0,190	0,496	-0,190	0,471	0,380	0,159	-0,270	0,319	-0,350	0,206	-0,050	0,862
	FPCB^c	0,260	0,362	0,170	0,535	-0,210	0,480	-0,080	0,773	0,250	0,383	0,220	0,422	-0,040	0,893	0,460	0,073	0,040	0,905	0,050	0,845
	FPS7^d	0,040	0,881	0,230	0,401	0,260	0,371	-0,050	0,862	0,390	0,169	-0,070	0,791	-0,420	0,134	0,160	0,559	0,180	0,531	-0,070	0,790
Stance Time [s]	SSP^a	0,040	0,899	0,250	0,343	-0,030	0,914	-0,380	0,142	-0,030	0,921	-0,130	0,636	0,330	0,226	0,110	0,683	0,170	0,535	-0,530	0,034
	FP^b	0,320	0,248	0,500	0,049	-0,410	0,133	-0,570	0,022	-0,530	0,043	-0,450	0,077	0,630	0,012	0,210	0,435	-0,060	0,834	-0,710	0,002
	FPCB^c	0,510	0,065	0,460	0,074	-0,530	0,053	-0,420	0,108	-0,390	0,172	-0,420	0,106	0,520	0,056	0,080	0,769	-0,010	0,970	-0,660	0,005
	FPS7^d	0,400	0,158	0,330	0,213	-0,480	0,085	-0,430	0,096	-0,380	0,175	-0,400	0,121	0,680	0,008	0,080	0,781	-0,240	0,407	-0,320	0,225
Double Support Time [s]	SSP^a	0,050	0,854	0,200	0,464	0,010	0,959	-0,340	0,191	-0,060	0,842	-0,090	0,738	0,300	0,286	0,140	0,596	0,270	0,339	-0,580	0,019
	FP^b	0,170	0,545	0,400	0,123	-0,360	0,193	-0,490	0,054	-0,480	0,068	-0,420	0,103	0,560	0,031	0,140	0,612	0,010	0,985	-0,640	0,008
	FPCB^c	0,540	0,045	0,370	0,161	-0,470	0,087	-0,310	0,240	-0,370	0,198	-0,440	0,087	0,580	0,028	0,000	0,987	-0,050	0,869	-0,640	0,007
	FPS7^d	0,220	0,459	0,300	0,262	-0,430	0,127	-0,430	0,095	-0,520	0,058	-0,420	0,103	0,530	0,050	0,070	0,807	-0,140	0,630	-0,280	0,300
Double Support Time Variability [s]	SSP^a	0,300	0,277	0,010	0,978	-0,260	0,341	-0,310	0,240	-0,300	0,282	0,010	0,963	-0,010	0,970	-0,120	0,671	0,210	0,449	0,160	0,562
	FP^b	0,230	0,405	0,270	0,315	-0,330	0,231	-0,470	0,068	-0,450	0,093	-0,300	0,263	-0,010	0,985	0,230	0,387	-0,280	0,309	-0,380	0,150
	FPCB^c	0,310	0,288	0,270	0,309	-0,420	0,134	-0,370	0,159	-0,480	0,080	-0,530	0,035	0,330	0,245	0,100	0,719	-0,050	0,857	-0,170	0,529
	FPS7^d	0,100	0,736	0,460	0,074	-0,400	0,160	-0,430	0,093	-0,620	0,019	-0,350	0,183	0,160	0,588	0,290	0,284	-0,450	0,108	-0,070	0,806
Swing Time [s]	SSP^a	-0,030	0,904	0,410	0,113	-0,060	0,839	-0,640	0,008	-0,020	0,952	-0,100	0,712	0,310	0,262	0,210	0,432	0,080	0,790	-0,280	0,294
	FP^b	0,550	0,035	0,740	< 0,001	-0,430	0,107	-0,570	0,020	-0,400	0,135	-0,520	0,039	0,690	0,005	0,300	0,259	-0,420	0,123	-0,500	0,051
	FPCB^c	0,620	0,017	0,840	< 0,001	-0,580	0,031	-0,590	0,016	-0,390	0,172	-0,310	0,243	0,400	0,155	0,470	0,067	0,010	0,982	-0,460	0,075
	FPS7^d	0,670	0,008	0,620	0,010	-0,520	0,058	-0,400	0,124	-0,140	0,643	-0,210	0,443	0,560	0,037	0,240	0,372	-0,230	0,425	-0,310	0,244
Mean Speed [m/s]	SSP^a	-0,640	0,010	-0,660	0,006	0,770	< 0,001	0,730	0,001	0,610	0,015	0,570	0,021	-0,390	0,147	-0,150	0,581	0,080	0,765	0,600	0,013
	FP^b	-0,660	0,007	-0,700	0,003	0,800	< 0,001	0,600	0,013	0,710	0,003	0,720	0,002	-0,430	0,107	-0,260	0,328	0,110	0,707	0,750	< 0,001
	FPCB^c	-0,640	0,014	-0,740	< 0,001	0,700	0,005	0,580	0,018	0,480	0,083	0,700	0,002	-0,320	0,266	-0,160	0,566	0,010	0,970	0,650	0,007
	FPS7^d	-0,770	0,001	-0,640	0,007	0,820	< 0,001	0,690	0,003	0,510	0,061	0,610	0,013	-0,360	0,207	-0,250	0,348	0,360	0,206	0,730	0,001
Cadence [steps/min]	SSP^a	0,040	0,899	-0,080	0,781	-0,060	0,824	0,220	0,406	-0,080	0,765	0,050	0,863	-0,360	0,183	0,000	0,987	-0,140	0,628	0,440	0,091
	FP^b	-0,230	0,420	-0,450	0,082	0,300	0,280	0,490	0,054	0,350	0,204	0,390	0,134	-0,690	0,004	-0,140	0,604	0,090	0,746	0,500	0,049
	FPCB^c	-0,500	0,066	-0,350	0,184	0,490	0,073	0,230	0,384	0,370	0,195	0,360	0,175	-0,630	0,015	0,010	0,974	0,110	0,707	0,620	0,011
	FPS7^d	-0,290	0,322	-0,220	0,414	0,380	0,179	0,320	0,233	0,240	0,415	0,250	0,343	-0,610	0,021	-0,070	0,811	0,160	0,587	0,270	0,319
Step Length [m]	SSP^a	-0,569	0,027	-0,742	0,001	0,721	0,002	0,691	0,003	0,567	0,028	0,635	0,008	-0,109	0,488	-0,168	0,535	0,116	0,680	0,597	0,015
	FP^b	-0,594	0,020	-0,636	0,008	0,699	0,004	0,572	0,021	0,495	0,060	0,678	0,004	-0,125	0,658	-0,073	0,787	0,060	0,831	0,605	0,013
	FPCB^c	-0,568	0,034	-0,748	0,001	0,645	0,013	0,597	0,015	0,466	0,093	0,712	0,002	-0,113	0,702	-0,119	0,661	0,027	0,928	0,584	0,018
	FPS7^d	-0,525	0,054	-0,738	0,001	0,591	0,026	0,622	0,010	0,426	0,129	0,698	0,003	0,006	0,985	-0,223	0,405	0,210	0,472	0,620	0,010
(^a) Self-Selected Pace; (^b) Fast Pace; (^c) Fast Pace Checking-boxes; (^d) Fast pace with serial seven subtractions; (^e)Scale for the Assessment and Rating of Ataxia; (^f) Berg Balance Scale; (^g) Barthel Index; (^h) Falls Efficacy Scale – International; (ⁱ) Montreal Cognite Assessment; (r) Spearman's rank correlation coefficient; (p) p-value

Functionality, as assessed by BI, was directly correlated with gait speed and step length in the SCA10 group during single and dual tasks. However, BI correlations with mean speed was only observed in the SCA3 group under SSP and FP conditions. During dual tasks, double support time variability was negatively correlated with BI in the SCA10 group under the FPCB condition and in the SCA3 group under the FPS7 condition. In both clinical groups, gait speed and step length decreased along with the BBS score during all walking tests. Furthermore, the swing time in the SCA10 group showed a negative correlation with BBS under the SSP, FP, and FPCB conditions, which was not observed in the SCA3 group.

No correlations were found between FES-I and the STPs in the SCA10 group. However, in SCA3 patients, there were strong positive correlations between step, stride, stance, and swing times under the FP and FPS7 conditions and FES-I. Cadence under the FP, FPCB, and FPS7 conditions in the SCA3 group were inversely correlated with FES-I. Negative correlations were found between MoCA and stride, step, stance, and double support times under the SSP, FP, and FPCB conditions in the SCA10 group. Additionally, MoCA was strongly positively correlated with mean speed and step length for all gait conditions. No correlations were found between MoCA and STPs in the SCA3 group.

4.1 SCA Gait Profile

Compared to healthy controls, the SCA groups displayed an unstable STP profile characterized by slowness, short steps, and a high-variability gait profile (Fig. 1). A recent systematic review found a similar cerebellar gait profile, suggesting that the main presentations in ataxias are reduced mean speed, cadence, and step length, with increased variability parameters, step width, step time, stance phase, and double support phase [18]. Other studies have suggested that key STPs in ataxia are trunk range of motion, gait speed, double stance time, and swing time [37–39].

In our study, step and stride times were longer in the SCA10 group compared to controls only when evaluated under the FP condition. These STPs were also higher in SCA3 patients under the SSP condition, which indicates worse gait performance due to lower adaptability. It has been well-established that step and stride times are longer in ataxic patients than in healthy controls [37, 40]. In gait ataxia the step length is reduced to achieve better trunk stability, leading to lower gait speed in everyday life, which then reduces the step time [37, 41]. In addition to decreased speed, a reduced step length may serve as a compensatory mechanism for increased variability in cerebellar gait ataxia [37].

The longer swing times found in both SCA groups confirm previous results obtained under SSP conditions using a force platform and wearable IMUs attached to the lower back in SCA6 patients [40], although other studies have been unable to confirm this finding [42–44]. Longer swing times may indicate greater gait instability, as the contralateral foot is not supported over long periods.

In our study, both step time and double support time variabilities were increased, even under the SSP condition. This is in line with a study that found increased stride time variability in ataxic patients during fast- and slow-paced walking but not during SSP walking, suggesting that SSP walking is the safest walking condition for ataxic patients [31]. An observational study of asymptomatic SCA2 gene carriers has suggested that variability parameters are elevated (even in preclinical stages), indicating that variability may be an early biomarker in SCA [39]. High gait variability may be a direct consequence of damage to cerebellar motor automaticity pathways [37, 41, 45–46] and dysfunction in the cerebellar pacemaker may contribute to increased fluctuation in gait rhythm between steps [31, 46]. A peripheral reduction in intra-articular temporal coordination and imprecise adjustments due to impaired balance control have also been identified [40, 47].

Consistent with our data, previous studies have shown that cadence reduction can occur in ataxia [48–51] and may result from increased durations of stance and double-support phases [44]. In patients with Friedreich’s ataxia, stance-phase duration increases with higher speed gait, highlighting the dependence of cadence performance on walking speed [47].

In our study, the SCA3 group showed significant time asymmetry only under the FPCB condition. This raises the question of whether there is an asymmetrical impairment of the cerebellar attentional pathways and whether interference from a dual-motor task contributes to worsening gait quality in patients with SCA3 due to asymmetry. Gait asymmetry is associated with increased variability, loss of balance, and gait instability [52] and has been previously observed in the SSP condition in SCA6 patients [51].

4.2 Gait speed ratios and reductions in speed gain capacity

Mean speed was the most relevant STP evaluated in our study (Fig. 1). Gait speed reduction has been demonstrated in the majority of studies evaluating gait parameters in ataxic patients, suggesting that decreased speed is a poor gait quality indicator in ataxias [37, 39–40, 42, 50–51, 53].

The gait speed ratio between the SSP and FP conditions (GS ratio SSP/FP) was significantly higher in the SCA groups than in the controls (table 3). This finding suggests that, even when walking at the fastest possible pace (FP), patients with SCA could not achieve the same proportional speed increase compared to healthy subjects. Thus, there was a reduction in speed gain capacity in both clinical groups, with no difference between them. Schniepp et al. found that, in patients with gait ataxia, there is a reduction in the maximum speed adopted during fast walking, which occurs, to a lesser extent, during preferred walking. This could be explained by a speed-dependent insufficiency of the cerebellar pacemaker region in ataxic patients, which is less evident during preferred gait [31]. In addition, reducing gait speed would be an attempt to better integrate proprioceptive feedback with ideal motor memory stored in the lateral cerebellum and central executive network [54].

Therefore, in ataxia, there is less adaptive capacity to increase gait speed through environmental disturbances, which can worsen gait quality and increase the risk of falls [53, 55]. Other studies have shown that during fast walking, double support time [47], stride time [46], and stride length variabilities [46] increases, indicating a decline in gait quality as faster gait patterns are adopted in ataxia. In addition, high variability in gait speed during FP walking has been associated with increased self-reported falls in a recent study [56].

4.3 DTC and cognitive cerebellar pathways

The GS-FPCB-DTC data suggest that patients with SCA10 may experience the effects of inattention and executive dysfunction on gait performance, which, surprisingly, was not apparent in the SCA3 group (table 3). Few studies have assessed STPs in ataxic patients using dual-task protocols and wearables. A recent study conducted in patients with a history of cerebellar lesions found decreasing step length, slower gait speed, increased step time, and cadence reduction under dual-task conditions [57]. Another study evaluated the cognitive dual-task effect on Fragile X-associated Tremor/Ataxia Syndrome (FXTAS) and found a higher DTC in men with FXTAS compared to healthy subjects and a positive correlation between stride velocity variability and the risk of falls [56].

The cerebellum plays a role in the attentional shift from a voluntary motor task to an automatic one. This affects gait when external environmental factors act as potential distractors, reducing gait quality and increasing the risk of falls [58]. Working memory impairment has been linked to increasing STP variability in patients with cerebellar lesions evaluated with a cognitive, tandem gait, dual-task protocol [59]. When attention is divided, motor tasks are prioritized to maintain balance, as the lateral cerebellar hemisphere is involved in pathways responsible for working memory and visuomotor control [60]. Therefore, deficits in the integration of the default mode and salience networks that occur between the neocerebellum, the pontine nuclei, the thalamus, the dorsal anterior cingulate cortex, and dorsolateral prefrontal cortex could explain poor performance on dual tasks [61].

While cognitive decline in SCA3 patients is well documented [62–64], few studies have evaluated cognition in SCA10 patients [65–66], although there is evidence suggesting impairment in short-term visual and visuospatial memory, reduced verbal memory storage capacity, and executive dysfunction in these patients [67]. A magnetic resonance imaging, tensor diffusion analysis-based study evaluated SCA10 patients and found demyelination patterns in subcortical pathways, including the parietal region of the superior longitudinal fasciculus and the cingulate fasciculus, both of which are involved in attention and executive functions such as working memory [68]. Although correlations have been found between STPs and MoCA results in SCA10 patients, the influence of cognitive impairment on STPs needs to be verified [69–71].

4.4 Correlations between STPs and clinical scales

Three STPs were strongly correlated with clinical scales: mean speed, swing time, and step length. SCA patients who presented with a slower short-step gait and spent more time in the swing phase also had a more advanced disease stage (SARA), worse static and dynamic balance in daily living activities, a higher risk of falls (BBS), and lower independence in the context of basic activities of daily living (BI). Correlations between gait speed and SARA have also been found in SCA6 and SCA31 patients [17]. Moreover, one longitudinal study evaluating STPs followed SCA3 patients for one year and found that disease progression in SARA was correlated with stride length, lateral sway, and slow gait speed [72].

Our study found a negative correlation between step time variability and double support time variability with BI in SCA3 (FPS7) and SCA10 (FPCB) groups. These results suggest that SCA patients with lower functionality exhibit a more unstable and variable gait pattern during dual-task performance. High variability has already been correlated with an increased risk of falls in ataxic patients [55, 73]. Caliandro et al. also described positive correlations between step length variability, number of falls per year, and SARA [53].

In the SCA10 group, step, stride, and stance times during FP walking were negatively correlated with BBS. Moreover, FES-I was correlated only in SCA3 patients with lower gait cadence and longer swing, stride, step, and stance times when assessed under the FP and FPS7 conditions. Furthermore, dual tasks contributed to an increased fear of falling and gait insecurity in SCA3 patients, reflected in lower cadence. Gait automatism may worsen while engaged in fast walking, which leads to imbalance—thus explaining the higher risk of falls during fast walking. Moreover, attempting to walk at higher speeds increases insecurity when performing automatic movements in SCA3 patients, which may be a consequence of the aforementioned inability to gain gait speed [31, 55].

Both SCA groups had similar STP profiles when compared to controls. The STP profile for the clinical groups was as follows: increased step time, stride time, stance time, swing time, double support time, step time variability, double support time variability, and decreased mean speed, cadence, and step length. Thus, based on our results, SCA3 and SCA10 patients present a highly variable, short-stepped, slow gait pattern. We also found higher GS ratio SSP/FP values in both SCA groups compared to controls, suggesting a low-speed gain capacity. Dual-tasking tended to worsen gait quality overall. The GS- FPCB-DTC was higher in the SCA10 group compared to controls, indicating worse performance on motor dual-task. Our results also suggest that step length, swing time, and gait speed may be good predictors for falling risk and disease progression. Specifically, a fear of falls was correlated with STPs in SCA3 patients, and a worse MoCA performance was correlated with STPs in SCA10 patients.

Ethics Approval

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by HC-UFPR’s Committee for Ethics in Human Research (CAAE 88366318.1.0000.0096)

Consent to Participate

All participants included in this study signed a voluntary informed consent form.

Author Contribution

Conceptualization: J.G., M.M., W.M., I.B., H.A.G.T.; Organization: I.B., C.H., M.M.; Clinical evaluations: I.B., G.K.S; Gait protocol data collection: I.B.; Raw data processing: C.H.; Interpretation of results: I.B.; Manuscript preparation: I.B.; Review and critique: W.M., R.P.M., M.M., H.A.G.T.

Data Availability

Data is provided within the manuscript

Schöggl J, Siegert S, Boltshauser E, Freilinger M, Schmidt WM. A De Novo Missense NPTX1 Variant in an Individual with Infantile-Onset Cerebellar Ataxia. Mov Disord. 2022;37:1774–6.
Coutelier M, Jacoupy M, Janer A, Renaud F, Auger N, Saripella GV, et al. NPTX1 mutations trigger endoplasmic reticulum stress and cause autosomal dominant cerebellar ataxia. Brain. 2022;145:1519–34.
De Castilhos RM, Furtado GV, Gheno TC, Schaeffer P, Russo A, Barsottini O, et al. Spinocerebellar ataxias in Brazil - Frequencies and modulating effects of related genes. Cerebellum. 2014;13:17–28.
Takiyama Y, Nishizawa M, Tanaka H, Kawashima S, Sakamoto H, Karube Y, et al. The gene for Machado–Joseph disease maps to human chromosome 14q. Nat Genet. 1993;4:300–4.
Stevanin G, Cancel G, Durr A, Chneiweiss H, Dubourg O, Weissenbach J, et al. The gene for spinal cerebellar ataxia 3 (SCA3) is located in a region of ~ 3 cM on chromosome 14q24.3-q32.2. Am J Hum Genet. 1995;56:193–201.
Moro A, Munhoz RP, Arruda WO, Raskin S, Moscovich M, Teive HAG. Spinocerebellar ataxia type 3: Subphenotypes in a cohort of brazilian patients. Arq Neuropsiquiatr. 2014;72:659–62.
Jacobi H, Bauer P, Giunti P, Labrum R, Sweeney MG, Charles P, et al. The natural history of spinocerebellar ataxia type 1, 2, 3, and 6: A 2-year follow-up study. Neurology. 2011;77:1035–41.
Matsuura T, Fang P, Pearson CE, Jayakar P, Ashizawa T, Roa BB, et al. Interruptions in the expanded ATTCT repeat of spinocerebellar ataxia type 10: Repeat purity as a disease modifier? Am J Hum Genet. 2006;78:125–9.
Raskin S, Ashizawa T, Teive HAG, Arruda WO, Fang P, Gao R, et al. Reduced penetrance in a Brazilian family with spinocerebellar ataxia type 10. Arch Neurol. 2007;64:591–4.
Teive HAG, Ashizawa T. Spinocerebellar ataxia type 10: From amerindians to latin americans. Curr Neurol Neurosci Rep. 2013;13:9–11.
Rasmussen A, Matsuura T, Ruano L, Yescas P, Ochoa A, Ashizawa T, et al. Clinical and genetic analysis of four Mexican families with spinocerebellar ataxia type 10. Ann Neurol. 2001;50:234–9.
Matsuura T, Ranum LPW, Volpini V, Pandolfo M, Sasaki H, Tashiro K, et al. Spinocerebellar ataxia type 10 is rare in populations other than Mexicans. Neurology. 2002;58:983–4.
Gheno TC, Furtado GV, Saute JAM, Donis KC, Fontanari AMV, Emmel VE, et al. Spinocerebellar ataxia type 10: common haplotype and disease progression rate in Peru and Brazil. Eur J Neurol. 2017;24:892–e36.
Véliz-Otani D, Cubas-Montecino D, Milla-Neyra K, Ashizawa T, Saraiva-Pereira ML, Jardim LB et al. Response to ATXN10 Microsatellite Distribution in a Peruvian Amerindian Population. Cerebellum. 2021.
Teive HAG, Munhoz RP, Raskin S, Arruda WO, de Paola L, Werneck LC, et al. Spinocerebellar ataxia type 10: Frequency of epilepsy in a large sample of Brazilian patients. Mov Disord. 2010;25:2875–8.
Ashizawa T, Xia G. Ataxia Articles from Continuum: Lifelong Learning in Neurology are provided here courtesy of American Academy of Neurology. Contin Mineap Minn. 2016;22:1208–26.
Matsushima A, Yoshida K, Genno H, Murata A, Matsuzawa S, Nakamura K, et al. Clinical assessment of standing and gait in ataxic patients using a triaxial accelerometer. Cerebellum Ataxias. 2015;2:1–7.
Buckley E, Mazzà C, McNeill A. A systematic review of the gait characteristics associated with Cerebellar Ataxia. Gait Posture. 2018;60:154–63.
Geritz J, Maetzold S, Steffen M, Pilotto A, Corrà MF, Moscovich M, et al. Motor, cognitive and mobility deficits in 1000 geriatric patients: protocol of a quantitative observational study before and after routine clinical geriatric treatment - The ComOn-study. BMC Geriatr. 2020;20:1–13.
Donath L, Faude O, Lichtenstein E, Nuesch C, Mündermann A. Validity and reliability of a portable gait analysis system for measuring spatiotemporal gait characteristics: Comparison to an instrumented treadmill. J Neuroeng Rehabil. 2016;13:1–9.
Nüesch C, Roos E, Pagenstert G, Mündermann A. Measuring joint kinematics of treadmill walking and running: Comparison between an inertial sensor based system and a camera-based system. J Biomech. 2017. http://dx.doi.org/10.1016/j.jbiomech.2017.03.015.
Bettecken K, Bernhard F, Sartor J, Hobert MA, Hofmann M, Gladow T, et al. No relevant association of kinematic gait parameters with Health-related Quality of Life in Parkinson’s disease. PLoS ONE. 2017;12:1–11.
Pham MH, Warmerdam E, Elshehabi M, Schlenstedt C, Bergeest LM, Heller M, et al. Validation of a lower back wearable-based sit-to-stand and stand-to-sit algorithm for patients with Parkinson’s disease and older adults in a home-like environment. Front Neurol. 2018;9:1–11.
Pham MH, Elshehabi M, Haertner L, Del Din S, Srulijes K, Heger T, et al. Validation of a step detection algorithm during straight walking and turning in Patients with Parkinson’s disease and older adults using an inertial measurement unit at the lower back. Front Neurol. 2017;8:1–9.
Metzger FG, Hobert MA, Ehlis AC, Hasmann SE, Hahn T, Eschweiler GW, et al. Dual tasking for the differentiation between depression and mild cognitive impairment. Front Aging Neurosci. 2016;8:1–9.
Hobert MA, Meyer SI, Hasmann SE, Metzger FG, Suenkel U, Eschweiler GW, et al. Gait is associated with cognitive flexibility: A dual-tasking study in healthy older people. Front Aging Neurosci. 2017;9:1–9.
Salkovic D, Hobert MA, Bellut C, Funer F, Renno S, Haertner L, et al. Evidence for a Selectively Regulated Prioritization Shift Depending on Walking Situations in Older Adults. Front Aging Neurosci. 2017;9:1–9.
Del Din S, Elshehabi M, Galna B, Hobert MA, Warmerdam E, Suenkel U, et al. Gait analysis with wearables predicts conversion to parkinson disease. Ann Neurol. 2019;86:357–67.
Kressig RW, Beauchet O, Gaitrite E, Group N. Guidelines for clinical applications of spatio-temporal. Aging Clin Exp Res. 2005;18:174–6.
Bock O, Engelhard K, Guardiera P, Allmer H, Kleinert J. Gerontechnology and human cognition. IEEE Eng Med Biol Mag. 2008;27:23–8.
Schniepp R, Wuehr M, Neuhaeusser M, Kamenova M, Dimitriadis K, Klopstock T, et al. Locomotion speed determines gait variability in cerebellar ataxia and vestibular failure. Mov Disord. 2012;27:125–31.
Braga-Neto P, Godeiro-Junior C, Dutra LA, Pedroso JL, Barsottini OGP. Translation and validation into Brazilian version of the Scale of the Assessment and Rating of Ataxia (SARA). Arq Neuropsiquiatr. 2010;68:228–30.
Camargos FFO, Dias RC, Dias JMD, Freire MTF. Cross-cultural adaptation and evaluation of the psychometric properties of the Falls Efficacy Scale-International Among Elderly Brazilians (FES-I-BRAZIL). Rev Bras Fisioter Sao Carlos Sao Paulo Braz. 2010;14:237–43.
Miyamoto ST, Lombardi I, Berg KO, Ramos LR, Natour J. Brazilian version of the Berg balance scale. Braz J Med Biol Res. 2004;37:1411–21.
Minosso JSM, Amendola F, Alvarenga MRM, de Oliveira MA. Validação, no Brasil, do Índice de Barthel em idosos atendidos em ambulatórios. Acta Paul Enferm. 2010;23:218–23.
Sarmento ALR. Apresentação e aplicabilidade da versão brasileira da MoCA (Montreal Cognitive Assessment)para rastreio de Comprometimento Cognitivo Leve. 2009. 82 f. Dissertação (Mestrado) - Escola Paulista de Medicina, Universidade Federal de São Paulo (UNIFESP), São Paulo, 2009.
Serrao M, Pierelli F, Ranavolo A, Draicchio F, Conte C, Don R, et al. Gait pattern in inherited cerebellar ataxias. Cerebellum. 2012;11:194–211.
Bunn LM, Marsden JF, Giunti P, Day BL. Stance instability in spinocerebellar ataxia type 6. Mov Disord. 2013;28:510–6.
Velázquez-Pérez L, Rodriguez-Labrada R, González-Garcés Y, Arrufat-Pie E, Torres-Vega R, Medrano-Montero J, et al. Prodromal Spinocerebellar Ataxia Type 2 Subjects Have Quantifiable Gait and Postural Sway Deficits. Mov Disord. 2021;36:471–80.
Hickey A, Gunn E, Alcock L, Del Din S, Godfrey A, Rochester L, et al. Validity of a wearable accelerometer to quantify gait in spinocerebellar ataxia type 6. Physiol Meas. 2016;37:N105–17.
Palliyath S, Hallett M, Thomas SL, Lebiedowska MK. Gait in patients with cerebellar ataxia. Mov Disord. 1998;13:958–64.
Caliandro P, Serrao M, Padua L, Silvestri G, Iacovelli C, Simbolotti C, et al. Prefrontal cortex as a compensatory network in ataxic gait: A correlation study between cortical activity and gait parameters. Restor Neurol Neurosci. 2015;33:177–87.
Gouelle A, Mégrot F, Presedo A, Husson I, Yelnik A, Penneçot GF. The Gait Variability Index: A new way to quantify fluctuation magnitude of spatiotemporal parameters during gait. Gait Posture. 2013;38:461–5.
Stephenson J, Zesiewicz T, Gooch C, Wecker L, Sullivan K, Jahan I, et al. Gait and balance in adults with Friedreich’s ataxia. Gait Posture. 2015;41:603–7.
Ilg W, Golla H, Thier P, Giese MA. Specific influences of cerebellar dysfunctions on gait. Brain. 2007;130:786–98.
Wuehr M, Schniepp R, Ilmberger J, Brandt T, Jahn K. Speed-dependent temporospatial gait variability and long-range correlations in cerebellar ataxia. Gait Posture. 2013;37:214–8.
Milne SC, Hocking DR, Georgiou-Karistianis N, Murphy A, Delatycki MB, Corben LA. Sensitivity of Spatiotemporal Gait Parameters in Measuring Disease Severity in Friedreich Ataxia. Cerebellum. 2014;13:677–88.
Ebersbach G, Sojer M, Valldeoriola F, Wissel J, Müller J, Tolosa E, et al. Comparative analysis of gait in Parkinson’s disease, cerebellar ataxia and subcortical arteriosclerotic encephalopathy. Brain. 1999;122:1349–55.
Ienaga Y, Mitoma H, Kubota K, Morita S, Mizusawa H. Dynamic imbalance in gait ataxia. Characteristics of plantar pressure measurements. J Neurol Sci. 2006;246:53–7.
Matsushima A, Yoshida K, Genno H, Ikeda S. ichi. Principal component analysis for ataxic gait using a triaxial accelerometer. J NeuroEngineering Rehabil. 2017;14:1–7.
Rochester L, Galna B, Lord S, Mhiripiri D, Eglon G, Chinnery PF. Gait impairment precedes clinical symptoms in spinocerebellar ataxia type 6. Mov Disord. 2014;29:252–5.
Azadian E, Torbati HRT, Kakhki ARS, Farahpour N. The effect of dual task and executive training on pattern of gait in older adults with balance impairment: A Randomized controlled trial. Arch Gerontol Geriatr. 2016;62:83–9.
Caliandro P, Conte C, Iacovelli C, Tatarelli A, Castiglia SF, Reale G, et al. Exploring risk of falls and dynamic unbalance in cerebellar ataxia by inertial sensor assessment. Sens Switz. 2019;19:1–9.
Bares M, Lungu O, Liu T, Waechter T, Gomez CM, Ashe J. Impaired predictive motor timing in patients with cerebellar disorders. Exp Brain Res. 2007;180:355–65.
Schniepp R, Wuehr M, Schlick C, Huth S, Pradhan C, Dieterich M, et al. Increased gait variability is associated with the history of falls in patients with cerebellar ataxia. J Neurol. 2014;261:213–23.
O’Keefe JA, Guan J, Robertson E, Biskis A, Joyce J, Ouyang B, et al. The Effects of Dual Task Cognitive Interference and Fast-Paced Walking on Gait, Turns, and Falls in Men and Women with FXTAS. Cerebellum. 2021;20:212–21.
Sütçü G, Doğan M, Topuz S. Investigation of postural control and spatiotemporal parameters of gait during dual tasks in ataxic individuals. Neurol Sci. 2022;5943–9.
Jacobi H, Alfes J, Minnerop M, Konczak J, Klockgether T, Timmann D. Dual task effect on postural control in patients with degenerative cerebellar disorders. Cerebellum Ataxias. 2015;2:1–7.
Ilg W, Christensen A, Mueller OM, Goericke SL, Giese MA, Timmann D. Effects of cerebellar lesions on working memory interacting with motor tasks of different complexities. J Neurophysiol. 2013;110:2337–49.
Ilg W, Timmann D. Gait ataxia-specific cerebellar influences and their rehabilitation. Mov Disord. 2013;28:1566–75.
Habas C, Kamdar N, Nguyen D, Prater K, Beckmann CF, Menon V, et al. Distinct cerebellar contributions to intrinsic connectivity networks. J Neurosci. 2009;29:8586–94.
Schmahmann JD, Sherman JC. The cerebellar cognitive affective syndrome. Brain. 1998;121:561–79.
Bürk K, Globas C, Bösch S, Klockgether T, Zühlke C, Daum I, et al. Cognitive deficits in spinocerebellar ataxia type 1, 2, and 3. J Neurol. 2003;250:207–11.
Braga-Neto P, Pedroso JL, Alessi H, Dutra LA, Felício AC, Minett T, et al. Cerebellar cognitive affective syndrome in machado Joseph disease: Core clinical features. Cerebellum. 2012;11:549–56.
Moro A, Teive HAG. Comprometimento cognitivo na ataxia espinocerebelar do tipo 10. Dement e Neuropsychol. 2016;10:310–4.
Moro A, Munhoz RP, Moscovich M, Arruda WO, Raskin S, Silveira-Moriyama L, et al. Nonmotor Symptoms in Patients with Spinocerebellar Ataxia Type 10. Cerebellum. 2017;16:938–44.
Chirino-Pérez A, Vaca-Palomares I, Torres DL, Hernandez-Castillo CR, Diaz R, Ramirez-Garcia G et al. Cognitive Impairments in Spinocerebellar Ataxia Type 10 and Their Relation to Cortical Thickness. Mov Disord. 2021;1–13.
Meira AT, Arruda WO, Ono SE, Franklin GL, de Carvalho Neto A, Raskin S, et al. Analysis of diffusion tensor parameters in spinocerebellar ataxia type 3 and type 10 patients. Parkinsonism Relat Disord. 2020;78:73–8.
Dumurgier J, Artaud F, Touraine C, Rouaud O, Tavernier B, Dufouil C, et al. Gait speed and decline in gait speed as predictors of incident dementia. J Gerontol - Ser Biol Sci Med Sci. 2017;72:655–61.
Verghese J, Holtzer R, Lipton RB, Wang C. Quantitative gait markers and incident fall risk in older adults. J Gerontol - Ser Biol Sci Med Sci. 2009;64:896–901.
Verghese J, Wang C, Lipton RB, Holtzer R, Xue X. Quantitative gait dysfunction and risk of cognitive decline and dementia. J Neurol Neurosurg Psychiatry. 2007;78:929–35.
Ilg W, Müller B, Faber J, van Gaalen J, Hengel H, Vogt IR, et al. Digital Gait Biomarkers Allow to Capture 1-Year Longitudinal Change in Spinocerebellar Ataxia Type 3. Mov Disord. 2022;37:2295–301.
Schniepp R, Schlick C, Pradhan C, Dieterich M, Brandt T, Jahn K, et al. The interrelationship between disease severity, dynamic stability, and falls in cerebellar ataxia. J Neurol. 2016;263:1409–17.

No competing interests reported.

Spatiotemporal gait analysis of patients with spinocerebellar ataxia types 3 and 10 using Inertial Measurement Units: a comparative study

Status:

Journal Publication

Version 1

Abstract

Figures

1. Introduction

2. Materials and Methods