Friedreich ataxia (FA) is a progressive, autosomal recessive, multisystem neurodegenerative disorder with an onset usually in early adolescence. Although rare, it is the most commonly inherited ataxia.1 FA is associated with a number of systemic abnormalities including cardiomyopathy, dysarthria, scoliosis, pes cavus, and, to a lesser, extent diabetes mellitus and visual dysfunction. The hallmark of FA is progressive gait and limb ataxia resulting from spinocerebellar and dorsal column degeneration.2 The relatively recent evolution of understanding of the genetic basis and pathogenesis of the disease has resulted in hopes for the development of new therapies to slow the progression of and ultimately cure this fatal disease.3
Because of the multisystem involvement and associated impairment of locomotor function, gait parameters such as velocity, cadence, stride length, and duration of swing and stance phase during the gait cycle would be expected to be altered by the disease process. To date, there has been neither a systematic analysis of gait parameters in subjects with a diagnosis of FA4 nor a study that explicitly investigates relationships between these gait parameters and maturation. For clinicians who may have occasion to examine a child with a diagnosis of FA, an understanding of the relationship between disease severity and locomotor status can lead to a more well-informed plan of care.
In children with a diagnosis of cerebral palsy, it has been shown that as general motor function decreases, there is an associated decrease in locomotor function.5–7 Similar information has not been adequately reported in those with FA and may be useful in assessing the impact of disease progression, the need for mobility assistance, and the impact of intervention. The Friedreich Ataxia Rating Scale (FARS) is a recently developed scale used to characterize disease severity.8 Correlations between the FARS and disease duration have been established,8,9 as has the relationship between the FARS and other functional outcome measures including the Functional Independence Measure and the Modified Barthel Index.10 Investigators have found significant relationship among overall function, the ability to perform activities of daily living, and the FARS9,10 This study describes relationships between the FARS and basic gait parameters.
Not until recently has FA disease severity been characterized by locomotor status.11 In a study of adults with FA, Wilson et al,11 divided subjects into subgroups based on FARS score and defined groups as ambulatory, transitioning from walking to wheelchair use, or exclusively using a wheelchair. This classification of locomotor status provides a clinically meaningful categorization to researchers and clinicians. Similar categorizations are described in this analysis of children and adolescents. Additionally, correlations among the FARS, its subscales, and locomotor status elucidate the relationship of disease severity, impairment, and general locomotor function.
The purposes of this study were to quantify gait parameters in young children and adolescents with a diagnosis of FA, to characterize the relationship between locomotor status and disease severity as measured by the FARS, and to determine whether FARS scores can discriminate between subjects who are able to walk independently and those who require assistance.
This study was a descriptive study conducted as part of a double-blind, placebo-controlled trial examining the safety and efficacy of the drug idebenone administered to children and adolescents with a diagnosis of FA. It was conducted by the National Institute of Neurological Disorders and Stroke (NINDS) with the approval of the NINDS Institutional Review Board. This trial was registered with Clinicaltrials.gov (NCT00229632) and conducted at the National Institutes of Health. Data describing gait characteristics of the subjects that are included in this study were collected as part of a baseline assessment.
Forty-eight children (aged 5-12 years) and adolescents (aged 13-17 years) were enrolled in the study after informed consent was provided by their parents or guardians and subjects provided assent. Inclusion criteria included the ability to walk 25 feet with or without an assistance device, genetically confirmed FA, evidence of neurological impairment, between the ages of 9 and 17 years, and weight between 30 and 80 kg. In addition, the subjects were not allowed to have been exposed to idebenone, coenzyme Q10, or other dietary supplements for a period of at least one month before enrollment. Nearly all subjects (96%) were white, which is reflective of the predominant ethnicity of those with a diagnosis of FA.2
Subjects underwent a comprehensive examination that included a complete history and neurological assessments including the FARS.8 The FARS is an ordinal rating of disease severity. Interrater reliability, construct validity, and sensitivity of the FARS to measure change in disease severity have been established.8,10 Higher FARS scores are associated with greater disability, with total FARS scores ranging from 0 (asymptomatic) to 117 (complete disability). There are six FARS subscales that measure bulbar and speech function, upper and lower extremity coordination, neuropathy, and stability. Each subscale was developed to capture impairments specific to FA. A single rater trained to perform the FARS administered the test to all subjects. During the test, subjects perform standardized tasks, which were either measured (eg, timed) and then converted to an ordinal score or rated on an ordinal scale.
Gait parameters were collected using the Stride Analyzer computerized system (B&L Engineering, Tustin, CA). The Stride Analyzer is a microprocessor/PC system designed to record foot-floor contact data from footswitches and to calculate gait parameters from these data. Footswitches embedded in variously sized insoles were positioned at the heel, fifth metatarsal, first metatarsal, and great toe areas to indicate beginning and ending periods of foot contact. Footswitches were tested before each use to ensure operability. Test-retest reliability for gait speed measured by the Stride Analyzer is reported as good to excellent (intraclass correlation coefficients [2,1] = 0.88-0.98) in patients with neurological disorders.12,13 Ten intrasession serial measures of gait performance over a 150-minute period yields stable measurements based on time-series analysis.12
Subjects walked along a straight 10-m hallway at their customary walking speed; three trials of walking were recorded. Sufficient space was allowed for acceleration and deceleration before the start and after the end of the walkway. If subjects were unable to complete all three trials, data were averaged from the trials completed. For safety, use of a U-Step wheeled walker (Instep Mobility, Skokie, IL), orthosis, and physical assistance was allowed if needed. If a subject used an orthosis, the Stride Analyzer insoles were placed between the foot and the orthosis. Subjects were allowed to rest as needed after each trial. Parameters collected included gait velocity, cadence, stride length, gait cycle time, time in double-limb support, swing, and stance.
Analyses were performed using the Statistical Package for Social Science version 15.0 (SPSS, Inc., Chicago, IL). Descriptive statistics were used to summarize spatial and temporal parameters of gait. Velocity and stride length were normalized for height differences by dividing each by the subject’s height. t Tests were used to determine age, disease onset, and anthropometric differences between groups. Pearson correlation coefficients were computed to determine the strength of association between the total FARS score and gait parameters. Eta correlations were used to determine association between total FARS scores and subscales scores and the dichotomous variable of locomotor status (ie, independent walker and assisted walker). A Mann-Whitney U test of differences was applied to the highest significant correlation of variables measured to determine the ability of the FARS scores to discern differences between groups.
Subject Demographics and Locomotor Status
Forty-eight subjects were recruited to participate in the study; data from 38 subjects were included in the analysis. Gait data for 10 subjects were lost due to instrumentation error (n = 6) or inability of the subject to walk the required distance in the time required by the instrumentation procedures (n = 4). Data from 14 children and 24 adolescents were analyzed (38 total). The mean age of the subjects was 13 ± 2.3 years with a mean age at disease onset at 8 ± 3.1 years old. Mean weight and height were 47.2 ± 13.9 kg and 153.7 ± 12.9 cm, respectively. Body mass index, an estimate of body composition, was calculated and found to be within normal limits (19.6 ± 3.3) for the group. Previously established guidelines to categorize normal body mass index were used.14 Twenty-three individuals (60.5%) walked without support or assistance (group 1 = independent walkers). Fifteen subjects (39.5%) required assistance to walk (group 2 = assisted walkers). Assistance was defined as the need to reach out for intermittent support provided by a railing (n = 1) or the examiner’s hand (n = 1) or the use of a walker to walk safely (n = 13). No significant differences were found between the groups relative to age, height, disease onset, or anthropometrics; group demographics are given in Table 1. Six of the 38 subjects (16%) used some type of in-shoe orthotic or, in one case, a unilateral heel lift; in three cases, the orthotics provided support above the level of the malleolus.
Mean spatial and temporal measures of gait for all 38 subjects and the breakdown for independent (group 1) and assisted (group 2) walkers are listed in Table 2. Independent walkers had a significantly higher velocity and cadence, translating into a shorter gait cycle, compared with subjects who required assistance. A negative correlation between age and velocity and between age and stride length was observed (r = −0.416 and −0.443, respectively; P < 0.01 for each). Adjusting velocity and stride length for height did not substantially change their relationship with age (r = −0.434 and −0.497, respectively; P < 0.01 for each).
Relationships Between FARS and Locomotor Status
The mean total FARS score for all subjects was 48 (±15.2; range, 16-80). The mean total FARS score for group 1 was 39.7 (±10.4; range, 16-63), and for group 2, it was 58.8 (±14.2; range, 34-80), indicating that, on average, the independent walker group had less severe disease. Correlations with total FARS scores and gait parameters resulted in r values of >0.5 for the parameters gait velocity, stride length, and cadence (r = 0.696, 0.667, and 0.537, respectively; P < 0.001 for each). Total FARS scores were significantly correlated with locomotor status (ç value r = 0.623; P < 0.01). The FARS score of 50 defined the threshold for independent versus assisted walker status; 20 of the 23 independent walkers (87%) had a FARS score of ≤50 points, whereas 11 of the 15 subjects (73%) in group 2 had a FARS score of ≥50.
FARS subscale scores were also significantly correlated with independent versus poorer locomotor status; ç values ranged from r = 0.380 (for the upper extremity coordination subscale) to r = 0.630 (for the stability [balance] subscale; Table 3). The stability subscale scores were most strongly associated with the dichotomous locomotor status groupings. The stability subscale score was capable of discriminating between independent walkers and those requiring assistance, as determined by the Mann-Whitney U test of differences calculation. The mean stability subscale score for group 1 was 10.7 ± 3.8, and for group 2, it was 18.1 ± 5.6. Stability subscale scores of ≤15 correctly categorized 20 of 23 subjects (87%) as independent walkers, whereas scores of ≥15 correctly categorized 10 of 15 subjects (67%) as assisted walkers. Box plots of mean scores and standard deviation of total FARS scores and the stability subscale score are illustrated in Figure 1.
Subjects in this study demonstrated a wide range of locomotor function and disease severity. Although FA usually is diagnosed in adolescents, subjects in this group had an onset of symptoms at a mean age of 8 years. This could be reflective of a more astute parent population. For instance, if there are older siblings with FA in the family, younger siblings may be diagnosed earlier due to earlier recognition of the symptoms or earlier proactive genetic testing. Alternatively, the fact that these individuals had a earlier mean age at onset than typical may indicate that they were more severely affected than the general population of individuals with a diagnosis of FA.11
Older subjects tended to have a slower gait velocity, longer gait cycles, and slower cadence. Although these individuals had normal body mass indexes compared with similarly aged children without disability, these subjects had a slower velocity (0.88 m/sec vs 1.24 m/sec) and shorter stride lengths (1.0 m vs 1.2 m) and took fewer steps per minute (97.0 vs 123.4 steps).15 The mean gait cycle time of 1.35 seconds per cycle was also slower than that of children without disability. The percentage of double-limb support during the gait cycle was also longer than normal, occurring on average for 33.3% of the gait cycle vs 20%.16 Four individuals were not included in this study because they were unable to complete the 10-m walk in the default time allowed by the Stride Analyzer. Had these four subjects been included, the group’s overall velocity would have been slower and other parameters may have indicated even poorer gait performance.
Maturation has a distinct effect on gait.17,18,19 Typically, as children age, their gait velocity increases; in children older than 8 to 10 years of age, this is mostly attributed to a greater height and leg length and thus longer stride length.18 For the children and adolescents with FA, negative correlations between age and velocity and between age and stride length were observed, indicating that older children with FA have more advanced disease, tend to walk more slowly, and have a shorter stride length. Therefore, the trend in these subjects was in the direction opposite that of children without disability,18,19 a finding that may be unexpected for clinicians unfamiliar with FA. Additionally, stride length has been demonstrated to be approximately 76% to 86% of total height, regardless of age, in the population without disability.17 In the group of independent walkers, stride length was 79% of height, whereas in the group categorized as assisted walkers, stride length was 61% of height.
Other studies have found similar adaptations of gait parameters in individuals with impairments.20,21 Barak et al20 reported gait adaptations among older people who have a history of falling. Gait compensations in those subjects were characterized by faster cadence and shorter stride length. Quantifying basic parameters and the associated trends can improve assessment of status at given intervals of maturation.
In a study by Siegel et al,21 the relationship between walking ability and lower extremity impairment, specifically strength, in children and adolescents with a diagnosis of idiopathic inflammatory myopathies was examined. Subjects were able to walk without an assistive device, were aged 5-18 years old, and were diagnosed at a mean age of 8 years. The FA subjects’ independent walker group had a gait velocity similar to that of the group with idiopathic inflammatory myopathies (1.06 m/sec vs 1.03 m/sec). Shorter stride lengths and longer gait cycle times contributed to the slower than normal gait in the idiopathic inflammatory myopathies group. Subjects in this FA study, by walking with a slower cadence and shorter stride length, increased the amount of time spent in the stance phase. This may reflect their poor ability to maintain balance while elevating and progressing the leg forward during walking. Many individuals with FA have a propulsive gait. This gait pattern may seem to be uncontrolled, and it may seem as if the individual throws the foot forward to “catch” him- or herself.
The ratio of time spent in the swing or stance phase of the gait cycle typically does not change with age. Children without disability and older individuals tend to spend 40% to 44% of the gait cycle in swing and 56% to 60% of the gait cycle in stance.16,17 Subjects of this study who were independent walkers spent less time in the swing phase of gait (34.9%) and more time in the stance phase (65.1%) than healthy individuals. The percentage of the gait cycle spent in double-limb support was also greater in the group of independent walkers with FA compared with individuals without disability (30.2% vs 20%).16 When this group was compared with the group with idiopathic inflammatory myopathies, the double-limb support and stance time was shorter in the idiopathic inflammatory myopathies group, suggesting that even with similar gait velocities, balance was a greater concern for the FA group. Those in the assisted walker group spent even more time in stance phase (69.2%) and double-limb support (38.3%). This may not only be due to poor balance but also because a walker is typically moved forward during the double-limb support phase of gait.
Although all subjects in this study were able to walk, the FARS scores reflected an array of disease severity. FARS scores >50 typically separated independent walkers (group 1) from those requiring assistance (group 2). There was a strong correlation between total FARS scores and the dichotomous subject groupings based on locomotor status (ç value r = 0.623; P < 0.01). This relationship suggests that the FARS scores may be a useful screening tool to identify those requiring an assistive device for walking.
Subscales of the FARS demonstrated significant correlations with locomotor status. The strongest correlation was with the upright stability subscale, which represents 36 of the 117 total points of the FARS score. This subscale includes an ordinal rating of balance activities and consists of static balance tasks (eg, standing with eyes open or eyes closed). Only two items of the subscale require walking (8/36 points). The observed correlation suggests a strong relationship between static balance and dynamic activity. Given that both the total FARS score and the upright stability subscale have a similar strength of association with locomotor status, the stability subscale may also be useful to help clinicians anticipate the need for an assistive device. Future research is necessary to more fully describe the sensitivity and specificity of the scales, especially the stability subscale. Perhaps because the stability subscale measures only one subset of impairment found in FA, it was less accurate in identifying those who needed an assistive device. Total FARS scores were able to correctly identify 73% of individuals in group 2, whereas the stability subscale identified 67% correctly. With a larger group of subjects, this difference may prove to be smaller.
As the disease progresses and gait worsens, assistive devices such as walkers can become necessary for facilitating safe, upright locomotor function, maintaining mobility, and preventing the deleterious effects of immobility.2,21 In a case report,22 the use of a specialized walker significantly improved the subject’s functional mobility. However, not all individuals with ataxic gait are appropriate candidates for walker use. Accordingly, the findings of this study should not be interpreted to mean that every individual scoring ≥50 points on the total FARS scale or ≥15 points on the stability subscale is an appropriate candidate for a walker. Individuals with ataxia occasionally use lateral stepping to maintain balance; walkers can interfere with this compensatory technique.22 Individuals may also be less safe with a walker if they tip the walkers easily or if poor lower extremity proprioception results in frequent accidental contact with the wheels or legs of the walker. It is because of this type of mobility dysfunction that direct clinical examination by a physical therapist is warranted. However, therapists interacting with other healthcare providers can use these findings to further educate colleagues about the benefits of the FARS as a screening tool for assistive device needs and as a tool to help identify the need for further rehabilitation evaluation. As FARS scores approach the identified threshold values, primary care providers can justify rehabilitation evaluation of gait, mobility needs, and fall risk, especially if a referral has not already been initiated to address other limitations.
To date, there has been no documentation of temporal and spatial gait parameters in individuals with FA.4 This study, which focused on children and adolescents with FA, is the largest report of its kind and adds to the literature by reporting gait parameters that are impairment-based measures of locomotor function. However, because of the relatively small sample size (n = 38) and restriction to subjects 17 years old or younger, the results may not generalize to the larger population of individuals with FA.
Publications related to rehabilitation outcomes for those with a diagnosis of FA are few. This description of gait parameters, which are continuous measures, offers a useful complement to the ordinal-based rating scales that exist. The results demonstrated that a 10-m walk, a simple test that is fairly easily completed by the subjects, can be used to provide meaningful data. Analysis of gait parameters can help quantify changes in gait performance associated with disease severity. As observed in other populations with mobility deficits, gait parameters in those with FA are consistent with impairments and help characterize functional deficits.
In this study, subjects exhibited slower gait velocity and longer gait cycles compared with age-matched normative data. Inverse relationships between age and velocity and between age and stride length was observed, which is contrary to trends in age-matched individuals without disability. Knowledge of these trends and of changes in gait performance and associated outcomes enhances the development of the plan of care. Evidence of significant correlations between the FARS scores and locomotor status can help practitioners anticipate assistive device needs and support or justify the decision for referral to a rehabilitation specialist. Thus, these findings are intended to educate physicians and therapists who may have limited experience with FA and thereby improve the quality of care. Each advance in our understanding of these concepts can improve the ability to assess the efficacy of future rehabilitation and pharmacological interventions.
The authors thank Willie Ching, PT, NCS, and Angela Baker, NP, for their help in data collection.
1. Schols L, Amoiridis G, Przuntek H, et al. Friedreich’s ataxia revision of the phenotype according to molecular genetics. Brain.
2. Pandolfo M. Friedreich ataxia. Arch Neurol.
3. Delatycki MB, Ioannou PA, Churchyard AJ. Friedreich ataxia: from genes to therapies? Med J Aust.
4. Goulipian C, Bensoussan L, Viton JM, et al. Orthopedic shoes improve gait in Friedreich’s ataxia: a clinical and quantified case study. Eur J Phys Rehabil Med.
5. Damiano DL, Abel MF. Relation of gait analysis to gross motor function in cerebral palsy. Dev Med Child Neurol.
6. Drouin LM, Malouin F, Richards CL, et al. Correlation between the gross motor function measure scores and gait spatiotemporal measures in children with neurological impairments. Dev Med Child Neurol.
7. McDowell BC, Kerr C, Parkes J, et al. Validity of a 1 minute walk test for children with cerebral palsy. Dev Med Child Neurol.
8. Subramony SH, May W, Lynch DR, et al. Measuring Friedreich ataxia: interrater reliability of a neurologic rating scale. Neurology.
9. Lynch DR, Farmer JM, Wilson RL, et al. Performance measures in Friedreich ataxia: potential utility as clinical outcome tools. Mov Dis.
10. Fahey MC, Corben L, Collins V, et al. How is disease progress in Friedreich’s ataxia best measured? A study of four rating scales. J Neurol Neurosurg Psychiatr.
11. Wilson CL, Fahey MC, Corben LA, et al. Quality of life in Friedreich ataxia: what clinical, social and demographic factors are important? Eur J Neurol.
12. Morris ME, Matyas TA, Iansek R, et al. Temporal stability of gait in Parkinson’s disease. Phys Ther.
13. Hill KD, Goldie PA, Baker PA, et al. Retest reliability of the temporal and distance characteristics of hemiplegic gait using a footswitch system. Arch Phys Med Rehabil.
14. Bellizzi MC, Dietz WH. Workshop on childhood obesity: summary of the discussion. Am J Clin Nutr.
15. Oberg T, Karsznia A, Oberg K. Basic gait parameters: reference data for normal subjects, 10-79 years of age. J Rehabil Res Dev.
16. Perry J. Gait Analysis: Normal and Pathological Function.
Thorofare, NJ: Slack; 1992.
17. Beck RJ, Andriacchi TP, Kuo KN, et al. Changes in the gait patterns of growing children. J Bone Joint Surg Am.
18. Norlin R, Odenrick P, Sandlund B. Development of gait in the normal child. J Pediatr Orthop.
19. Wheelwright EF, Minns RA, Law HT, et al. Temporal and spatial parameters of gait in children. I. Normal control data. Dev Med Child Neuro.
20. Barak Y, Wagenaar RC, Holt KG. Gait characteristics of elderly people with a history of falls: A dynamic approach. Phys Ther.
21. Siegel KL, Hicks JE, Koziol DE, et al. Walking ability and its relationship to lower-extremity muscle strength in children with idiopathic inflammatory myopathies. Arch Phys Med Rehabil.
22. Harris-Love MO, Siegel KL, Paul S, et al. Rehabilitation management of Friedreich ataxia: lower extremity force-control variability and gait performance. Neurorehabil Repair.
23. Bateni H, Heung E, Zettel J, et al. Can use of walkers or canes impede lateral compensatory stepping movements? Gait Posture.
Keywords:© 2009 Academy of Neurologic Physical Therapy, APTA
degenerative disease; walking; ataxia