Relationship between clinical outcome measurements and muscle thickness in cerebral palsy

2.1 Participants

Nineteen children with CP (10 boys, mean age, 8.8 ± 1.8 years) were recruited from rehabilitation hospitals or community-based pediatric rehabilitation centers. This study conforms to the ethical standards of the Declaration of Helsinki. All the participants were notified of the purpose and experimental procedure of the study, and informed consent was obtained before participation (No. JHIRB-2020-01). The inclusion criteria included (1) children with hemiparetic or quadriplegic CP, (2) age ranging from 6 to 11 years, and (3) ability to follow instructions. The exclusion criteria included (1) history of musculoskeletal impairments, any surgical procedure within the last 6 months, or any medication influencing the experimental test. Table 1 presents the participants’ demographic characteristics.

The participants’ parents or caregivers completed a demographic and pre-health questionnaire concerning age, current medication, medical history, and other health issues. A trained physiotherapist at the children’s rehabilitation center consistently performed the tests. All the tests were conducted by the same investigator to improve the internal validity of the measurements.

2.3 ICF body structure domain-based ultrasound muscle thickness

Ultrasound imaging measurements were used to determine the muscle thickness and selective motor control as the structure domain variable measures in the more affected upper limb. Ultrasound sonography (Medison Co. Ltd., Seoul, South Korea) was performed with an 8-MHz linear array transducer to assess the muscle thickness of the triceps brachialis and extensor carpi radialis during maximal voluntary isometric contractions (MVIC) [11]. For extensor carpi radialis imaging, the child was placed in the supine position with the head and neck in a neutral position. The shoulder was abducted to 30 ∘ and internally rotated with full elbow extension and forearm pronation. The head of the radius was palpated and marked as the reference point. Exploratory scanning was first performed by moving the transducer to align transversally and moved 2 cm distally from the tip of the radial head until the sharpest cross-sectional area of the entire extensor carpi radialis was captured [12]. The cross-sectional area image was acquired during the resting and maximal isometric contraction conditions as the child exerted resistance to wrist extension and radial deviation movement against the therapist. For triceps brachialis scanning, the child was placed in the prone position with the head positioned on the testing side and the elbow flexed and abducted to 90 ∘ hanging over the side of the bed. First, the olecranon process was identified, and the probe was moved in a transverse and cephalic manner 10 cm from that point to capture the largest cross-sectional area of the entire triceps brachialis [12]. The cross-sectional area image was acquired during the resting and maximal isometric contraction conditions as the child exerted resistance to the therapist’s elbow extension movement. Although the cross-sectional areas of the bilateral triceps brachialis and extensor carpi radialis were acquired, only the cross-sectional area data recorded from the more affected side were used for further analysis. The MVIC of each muscle was consistently measured and monitored using a dynamometer to avoid any testing variability associated with the three repeated measurements.

2.4 ICF activity domain-based clinical motor function tests

The clinical motor and activity function tests included the quality of upper extremity skills test (QUEST) and Functional Independence Measure for Children (WeeFIM). QUEST is a criterion-referenced measurement tool for motor behaviors of the upper extremity. It examines the quality of upper extremity fine motor behaviors in four domains: dissociated movement (19 items with one level), grasp (six items), weight bearing, and protective extension [13]. The response criteria included “yes (2 points),” “no (1 point),” or “not tested (1 point).” The range of the initial grading scores was 50–100 and was standardized to range from zero (or below zero in the grasp section) to 100 using the formula while factoring for the not-tested items. The total test score was computed as the sum of all the domain scores divided by the number of tested domains.

The Wolf motor function test (WMFT) quantifies upper extremity motor ability through timed and functional tasks [14]. The WMFT consisted of 21 items. The first 6 items involve timed functional tasks, items 7 and 14 are measures of strength, and the remaining 9 items consist of analyzing movement quality when completing various tasks [15]. The examiner should test the less affected upper extremity followed by the most affected side. The following items should be performed as quickly as possible, truncated at 120 seconds. The reliability of the WMFT was r = 0.95 [16].

The WeeFIM is a criterion measure designed to assess the performance in functional skills [17]. The six WeeFIM subscales, comprising self-care, sphincter control, transfers, locomotion, communication, and social cognition, were scored on a 7-point scale. The WeeFIM instrument consists of six subsets with a total of 18 measurement items. The subsets were categorized as self-care (six items), sphincter control (two items), transfers (three items), locomotion (two items), communication (two items), and social cognition (three items). Each measurement item of the subsets is scored on a scale of 1–7, where 1 indicates total assistance and 7 indicates complete independence. The score ranges from 18 (total dependence) to 126 (independence) [18].

2.5 ICF participation domain-based clinical motor function tests

The Jebsen-Taylor Hand Function Test (JTHFT) is a referenced measure and designed to examine the hand function associated with activities of daily living tasks and amount of time spent to complete each test item [19]. It contains seven timed subtests: writing, card turning, picking up small items, simulated feeding, stacking checkers, and picking up light and heavy cans [20]. Time data were objectively measured using a stopwatch. The score ranges from 0 (total dependence) to 105 (independence) [20]. The validity of the JTHFT was r = 0.60 to 0.99 and validity was r = - 0.64 to - 0.69 [21].

2.6 Statistical analysis

Statistical data are expressed as mean and standard deviation. We used the G-Power software (G-Power version 3.1.5) to assess the sample size based on our previous pilot study which yielded 19 participants, computing from the effect size (Eta squared, η 2 = 0.6) and power ( 1 - β = 0.8) for muscle thickness. The statistical tests included the parametric Pearson correlation coefficient in determining multiple relationships between the body structure/function domain, the body activity domain, and participants’ domain variables. SPSS for Windows (version 26.0; IBM Corp., Armonk, NY, USA) was used, with the significance level set at α = 0.05.

3.1 Relationship between muscle thickness and activity

Pearson’s correlation analysis showed a moderate relationship ( r = 0.667, 0.05) between triceps brachii muscle thickness and WeeFIM, supporting that body structure was associated with the activity domain. The other variables were not correlated (Table 2).

Pearson’s correlation analysis showed moderate correlations between the JTHFT and ultrasound triceps brachii-affected side rest ( r = - 0.484, 0.05), between the JTHFT-affected side and ultrasound triceps brachii-affected side max ( r = - 0.591, 0.01), and between the JTHFT-affected side and ultrasound extensor carpi radialis-affected side max ( r = - 0.490, 0.05). However, the other variables were not correlated (Table 2).

3.3 Relationship between activity and participation variables

Spearman’s rank order analysis revealed a strong relationship between the WMFT side (sec) and JTHFT-affected side ( r = 0.893, 0.01). A strong relationship was observed between WMFT-affected side (ave) and JTHFT-affected side ( r = - 0.638, 0.01). A moderate relationship was noted between QUEST and WeeFIM, which was weak on the JTHFT-affected side ( r = - 0.338, 0.01) (Table 2).