**1. Introduction**

Cerebral palsy (CP) is a non-progressive upper motor neuron lesion where common motor disability occurs. Other manifestations of CP, such as loss of posture control and lack of movement, may also result in the development of musculoskeletal problem [1,2]. These problems arise because of spasticity and occurs in the hip, knee, and ankle joints [3]. Spasticity is a form of velocity-dependent resistance, or a motor disorder [4]. Children with spastic diplegic CP had greater spasticity at the ankles (more distal part) compared with the knees [5]. The ankle spasticity is related to the limited ankle joint movement [6]. In particular, the limited ankle joint movement in children with CP are closely related to gait and balance performance [7,8]. Therefore, the improvement of balance and gait ability regarding ankle function is one of the goals in rehabilitation for the musculoskeletal problems of children with CP [9,10].

Children with CP are observed to have a limited ankle ROM than typically developing children [11]. The limited ankle ROM of children with CP is also associated with higher tissue stiffness, increased reflexive torque of the gastrocnemius and soleus [11], and spasticity and weakness of the ankle joint [12]. The limited dorsiflexion ROM causes changes in contractile tissue as well as non-contractile tissue and is a common problem for children with CP [13,14].

A limited ankle ROM during gait and balance performance is an identified problem in children with CP [8,15,16]. On the other hand, an ankle ROM improvement may facilitate gait and balance performance [17,18]. For the optimal state of ankle dorsiflexion, it is essential to have a better understanding of the mechanism of limited ankle ROM, and to increase ankle ROM, not only osteokinematics but also arthrokinematics are required. Osteokinematics are simply movements of bones at the joints (flexion/extension, abduction/adduction, and internal rotation/external rotation), and arthrokinematics are small movements of bones at the joint surface (rolls, glides/slides, and spins).

Manual mobilization serves as an important component of neurorehabilitation to treat spasticity and limited ankle movements [19–22]. Manual mobilization is a passive movement performed to relieve pain, complete joint motion, and restore arthrokinematics rather than osteokinematics. Ankle joint mobilization can improve the associated limited ankle ROM characteristic of this condition, especially dorsiflexion [23,24]. Increased dorsi flexion ROM through joint mobilization improved the ankle kinematic changes during walking and [18] and static postural control [25]. These results seem to have a favorable e ffect on the sensorimotor function and arthrokinematic motion of the ankle [24,25].

Joint mobilization provides a well-suited intervention for ankle rehabilitation, but the e ffectiveness of a randomized control trial study in children with CP has not been confirmed. As part of manual therapy, there are spine manipulation studies for cerebral palsy [26,27]. Manipulation is a high-velocity (thrust) technique, while joint mobilization is a relatively safe technique with low-velocity techniques [28]. Manual therapy has been supported by several researchers that it can be used as a separate intervention method to treat secondary problems of musculoskeletal system in patients with spasticity (CP or stroke) [19–21,26,27,29–33].

Therefore, the purpose of this study is to investigate the e ffect of ankle joint mobilization to improve ankle ROM, standing balance, and gait in children with CP.

### **2. Materials and Methods**

### *2.1. Study Design*

This study was single blind, randomized controlled trial with two groups: a mobilization group (ankle joint mobilization) and control group (sham joint mobilization); individuals were randomly assigned using sequentially numbered, opaque-sealed envelopes. The base-line test was performed after obtaining a written informed consent from participants and their legal representatives. The base-line test is ankle dorsiflexion ROM, COP displacements, and gait function (timed up and go test, 10-m walk test). After the baseline-test, children with CP were randomly divided into 2 groups: the mobilization group (ankle joint mobilization, *n* = 16) and control group (sham joint mobilization, *n* = 16). The allocation ratio of the mobilization group/control group was 1:1. The gross motor skills of children with CP can be categorized into 5 di fferent levels using the gross motor function classification system (GMFCS). Since the GMFCS level can provide a confounding interpretation of the results, this study used a stratified block randomization form of the level of GMFCS (1 or 2) [34]. Children with CP were blinded to their treatment. The examiner was blinded to the group allocation. Only the physiotherapist who performed the joint mobilization knew the group to which the participants belonged to. The end-line test was performed after 6 weeks. This study was performed in accordance with the Declaration of Helsinki and research work was approved by the Institutional Review Board of Yong-in University (2-1040966-AB-N-01-20-1812-HSR-127-10).
