Structural Validity and Inter-rater Reliabitiliy of the Ataxia Trunk, Lower And Upper Extremity Scale (ATLAS)

Ataxia, derived from the Greek word ataxia meaning "disorder," is a neurological condition characterized by impaired coordination of oculomotor movements, upper limbs (UL), lower limbs (LL), and trunk, as well as dysarthria. Most of these symptoms lead to disturbances in postural control and gait. Ataxia may result from damage to the cerebellum, brainstem, and/or ascending pathways.

In addition to these motor impairments, vestibulocerebellar syndrome and cerebellar cognitive affective syndrome (also known as Schmahmann's syndrome) contribute to difficulties in regulating cognitive, motivational, and emotional states in affected individuals.

Ataxias are broadly classified into two main categories:

I) Hereditary or genetic ataxias, with an overall prevalence of approximately 2.7-38.35 per 100,000, include Friedreich's ataxia (autosomal recessive), other autosomal recessive ataxias (average prevalence of 3.3 per 100,000), and spinocerebellar ataxias (mostly autosomal dominant, average prevalence of 2.7 per 100,000).

II) Acquired ataxias, which may result from multiple sclerosis (MS; 100 per 100,000, with 10-50% showing signs of ataxia), stroke, traumatic brain injury (TBI), tumors, infections, vitamin deficiencies, or exposure to toxic substances such as alcohol, heavy metals, or medications.

Genetic ataxias affect men and women equally 1, whereas acquired ataxias show gender disparities depending on the underlying condition: MS affects approximately 1 male for every 4 females; stroke affects 1 male for every 1.28 females; TBI affects 3 males for every 1 female.

Across all types, ataxia affects approximately 1 in 10,000 individuals, making it a relatively rare condition. However, its impact on quality of life and employability is substantial, as it compromises mobility, independence, and the ability to perform daily activities. It increases reliance on healthcare professionals and reduces the quality of life of caregivers due to the growing burden. Ultimately, it also raises mortality risk. The average annual cost per patient was estimated at €18,776 in 2014. These psycho-socio-economic consequences justify scientific interest, particularly regarding treatment efficacy and appropriate assessment.

Treatments

Effective curative treatments are rare or nonexistent for most genetic ataxias. Only a few drugs, such as Riluzole and 4-Aminopyridine, may slightly slow disease progression or symptom expression. A Riluzole precursor, Triluzole, is currently undergoing phase III clinical trials. Management of acquired ataxias focuses on treating the underlying cause.

Non-pharmacological therapeutic options for both acquired and genetic ataxias include symptomatic and multidisciplinary rehabilitation, involving occupational therapy, intensive physiotherapy, and, where available, transcranial magnetic stimulation. Although the effectiveness of these interventions largely depends on the intensity and repetition of functional movements and balance exercises, interactive video game training has shown promise in improving motor abilities even in advanced stages.

Healthcare approaches are both restorative and compensatory. Compensatory aids such as movement decomposition, auditory and visual cues, viscoelastic resistance, computer adaptations (e.g., mouse speed), and walking aids (with or without weights) help maintain or perform activities important to the patient. Restorative approaches focus on repeating specific movements with or without biofeedback, functional movements, and postural control exercises in various environments and situations.

Traditional physical therapy aims to improve postural control and gait and includes physical (re-)conditioning, range-of-motion exercises, muscle strengthening, static and dynamic balance exercises in various positions (e.g., upright kneeling, quadruped, half-kneeling), spinal mobilization, and indoor/outdoor walking. Occupational therapy targets daily living activities such as hygiene, dressing, writing, and eating, and also includes coordination and balance exercises.

Nevertheless, ataxia remains the most treatment-resistant neurological symptom. Given the cerebellum's role in motor learning-particularly error-based learning and adaptation-individuals affected by ataxia often exhibit significant impairments in acquiring both simple and complex motor skills.

Consequently, it is essential to have assessment tools for ataxia that are not only valid and reliable but also sensitive to change. Typically, the evaluation of cerebellar motor syndrome includes five cardinal domains: gait and balance, upper limb coordination, lower limb coordination, speech clarity, and oculomotor control. To understand the differences among currently available scales, a description of the main instruments used to rate the severity of impairment is necessary.

Ataxia Rating Scales

A recent literature review identified 14 instruments and recommended the Friedreich's Ataxia Rating Scale (FARS), the International Cooperative Ataxia Rating Scale (ICARS), and the Scale for the Assessment and Rating of Ataxia (SARA) as the most appropriate tools for evaluating ataxia, excluding condition-specific scales such as the Unified Multiple System Atrophy Rating Scale (UMSARS). However, the four most widely used and studied scales are SARA, ICARS, FARS, and the Brief Ataxia Rating Scale (BARS).

These scales assess signs, symptoms, and the impact of cerebellar ataxia on daily activities, providing a clinical score. Originally developed by clinicians, they have been validated to reflect patients' lived experiences with cerebellar motor syndrome. A recent study confirmed the alignment between spontaneously reported symptoms by patients and caregivers and the concepts measured by these scales.

SARA (Scale for the Assessment and Rating of Ataxia) SARA is the most commonly used scale in both clinical and research settings worldwide. It was preferred over ICARS due to similar clinimetric properties-excellent validity and reliability, with no floor or ceiling effects-but with a shorter administration time. SARA consists of 8 items assessing gait, balance, upper and lower limb coordination, and speech articulation. Each item is scored on an ordinal scale, yielding a total score ranging from 0 (no ataxia) to 40 (severe ataxia).

Developed in 2006 for spinocerebellar ataxia, SARA has since been validated in populations with stroke, Friedreich's ataxia, multiple sclerosis (MS), and lysosomal storage disorders. It demonstrates high reliability: inter-rater (ICC: 0.921-0.98), intra-rater (ICC: 0.952-0.99), and test-retest (ICC: 0.90), with adequate internal consistency (Cronbach's α: 0.89-0.97 depending on population).

Correlations include:

MS: Expanded Disability Status Scale (rho = 0.557) Friedreich's Ataxia: FARS (r = 0.938) All ataxias: Barthel Index (r = -0.63) Stroke: Barthel Index (r = -0.792) In terms of sensitivity to change, for lysosomal storage disorders, an area under the curve (AUC) of 0.82 and a 1-point change is considered clinically significant. For spinocerebellar ataxias, the standardized response mean (SRM) is 0.50, with an average annual score change of 1.38/40 ± 2.8 SD.

A shortened version, the modified-functional SARA (f-SARA), was recently developed by a pharmaceutical group to provide a clinically meaningful measure of disease progression. It appears sensitive enough to detect significant changes over one year, with a 1-point intra-individual threshold . However, SARA remains more precise for tracking disease progression, as item reduction in f-SARA compromises its granularity .

ICARS (International Cooperative Ataxia Rating Scale) Developed in 1997, ICARS includes 19 items across four subcategories: posture and gait disturbances, limb ataxia, dysarthria, and oculomotor disorders. Most items are scored on an ordinal scale, except one binary item assessing saccadic dysmetria. The total score ranges from 0 (no ataxia) to 100 (severe ataxia) .

ICARS has been validated in various populations, including focal lesions, Friedreich's ataxia, chronic alcoholism, and spinocerebellar ataxias. It shows good validity and excellent reliability (inter-rater ICC: 0.95; intra-rater ICC: 0.97), but its 30-minute administration time limits routine use. ICARS is particularly effective in distinguishing mild from moderate ataxia in MS patients (AUC: 0.875) .

BARS (Brief Ataxia Rating Scale) Developed in 2009 as a shortened version of ICARS, BARS includes 5 items focusing on key motor domains: gait, heel-knee test, finger-nose test (scored 0-4), dysarthria, and oculomotor abnormalities. The total score ranges from 0 (no ataxia) to 19 (severe ataxia).

BARS is valid (correlation with SARA: r = 0.91), reliable (inter-rater ICC: 0.91), and sufficiently fast and accurate for clinical use, though it remains underutilized .

FARS (Friedreich's Ataxia Rating Scale)

FARS comprises five subdomains assessing:

(A) Bulbar function (B) Upper limb coordination (C) Lower limb coordination (D) Peripheral nervous system (E) Upright stability Only subdomains B, C, and E are specific to ataxia and overlap with SARA and ICARS. A shortened version of FARS also exists, maintaining excellent clinimetric properties.

Among all these scales, SARA is the most frequently used tool in clinical practice in French-speaking Switzerland, as it has undergone a transcultural validation process and is therefore available in French. However, physiotherapists involved in the rehabilitation of individuals with ataxia encounter limitations when using SARA to assess their patients. Specifically, SARA does not allow for the objectification of fluctuations that therapists observe in clinical settings. Moreover, some patients have reported that SARA scores do not meaningfully reflect their clinical condition .

One possible explanation is that SARA includes global activities such as gait and balance. In this population, improvements in activities are often contingent upon improvements in impairments. The first hypothesis is that it is necessary to improve functions or impairments-according to the terminology of the International Classification of Functioning, Disability and Health (ICF)-in order to impact activity levels. The goal was therefore to develop a scale that evaluates functions, enabling the detection of smaller variations in patients' motor abilities.

A second hypothesis is that more distal and fine motor activities may better capture changes. For example, in a population of individuals with ataxia, the Nine Hole Peg Test (9HPT), which assesses manual dexterity, was found to be more sensitive to change than SARA. Finally, SARA includes only a few global tests for upper and lower limbs, while the trunk is largely absent from the evaluated components. The objective of the new ATLAS scale was thus to address this gap by incorporating trunk assessment, evaluating limbs with greater precision, and including distal motor evaluation.

Another team of clinician-researchers recently sought to go beyond traditional time-based or task-completion measures used in SARA and ICARS, and to include trunk assessment. They developed the Comprehensive Coordination Scale (CCS), which evaluates both performance and movement quality. It consists of six tests assessing motor coordination of the trunk, upper limbs, lower limbs, or their combination: finger-to-nose test, arm-trunk coordination test, finger opposition, interlimb coordination, Lower Extremity Motor Coordination Test (LEMOCOT), and four-limb coordination.

ATLAS, developed by a local team, differs from CCS by including scoring for upper and lower trunk control across different planes, and by evaluating the shoulder, elbow, wrist, hip, knee, and ankle levels distinctly and specifically. Moreover, the final item in CCS-four-limb coordination-is not under cerebellar control and is therefore considered inappropriate for the objectives of our project.

Ataxia Trunk, Lower And

• Functional Independence Measure (FIM) [Time frame: Once, at day 1]
• Trunk Impairment Scale 2.0 (TIS 2.0) [Time frame: Once, at day 1]
• Scale for the Assessment and Rating of Ataxia (SARA) [Time frame: Once, at day 1]