*2.2. Participants*

A total of 32 children with CP were recruited from the Bundang Jesaeng general hospital in Gyeonggi do, Korea, February to May 2019. The inclusion criteria were (1) school-aged cerebral palsy children (8 to 14 years) (Mehraban et al. 2016); (2) a diagnosis of spastic diplegia; (3) gross motor function classification system (GMFCS) level I or II [35]; (4) hypomobility according to a 5-point posterior talar gliding test [36]; (5) ability to walk 10 m or more independently; and (6) children with CP to follow verbal directions. Exclusion criteria were (1) a history of selective dorsal rhizotomy and lower extremity orthopedic surgery; (2) botulinum toxin injections in leg muscles during the preceding year; and (3) visual disorder. The detailed study plan is depicted in Figure 1.

**Figure 1.** The flowchart of recruitment.

### *2.3. Sample Size Calculation*

G\*Power 3.19 (Heinrich Heine University, Dusseldorf, Germany) was used for sample size calculation. The sample size of this study was calculated based on the pilot test. Eight children with CP (four children in the joint mobilization group and four in the sham mobilization group) were involved to calculate the subjects needed for this study. The effect sizes of 1.30 (left) and 1.40 (right) were derived using the mean and standard deviation of dorsi flexion ROM in the supine position among the main outcomes. Based on the effect size of 1.30, input of a confidence level of 95%, and power of 80%, the total required sample size was 22. In this study, 32 participants were selected considering dropout. The initial participants were not included in the sample of 32.

### *2.4. Intervention Methods*

The period of the ankle intervention in this study was 6 weeks. All participants received ankle intervention 30 times (5 sessions per week) over a 6-week period. The mobilization and control groups received the neurodevelopment treatment (NDT) program for 6 successive weeks. The NDT program was performed to improve trunk control. The NDT program consisted of 20 min of trunk muscle exercise and upper extremity exercise (a reaching task of the upper limb for mobility, and trunk control for stability in the sagittal, coronal and transverse planes). The principles of NDT are trunk control in the sitting and standing positions [37]. The mobilization group additionally received (12 min) ankle joint mobilization. The control group additionally received (12 min) sham mobilization.

### *2.5. Ankle Joint Mobilization*

Ankle joint mobilization was performed to improve dorsiflexion ROM. Joint mobilization was performed by one physiotherapist certified in IMTA Maitland concept level 1, with over 10 years of neurodevelopmental treatments experience. After manual evaluation, ankle joint mobilization was carried out in the distal tibiofibular joint, talocrural joint, and subtalar joint. The manual evaluation for ankle joint mobilization procedure is depicted in Figure 2. Manual evaluation to determine the direction of the joint mobilization is as follows. The joint mobilization group recorded the direction of hypomobility among the hypermobility, normal and hypomobility, and performed large-amplitude, rhythmic oscillations (grade III) in the direction of the hypomobility [38]. In this study, except for the

distal tibiofibular joint, joint mobilization was applied in the AP direction, and participants received about 50 oscillations per set with 1 min of rest between sets. The rest time was 1 min and applied to both legs for a total of 12 min by applying joint mobilization to the opposite leg during a rest time of 1 min. Therefore, joint mobilization was applied to each of the three ankle joints for 4 min, and the oscillations technique was applied in 2 sets per one leg.

**Figure 2.** The ankle joint mobilization.

## *2.6. Sham Mobilization*

The sham mobilization visually resembles the joint mobilization. The sham mobilization seems to perform the same action as ankle joint mobilization, but only manual contact is performed because there is no direction of oscillations.
