Among musculoskeletal injuries, ankle sprains are common [1]. In addition, the recurrence rate of ankle sprain is high, and symptoms often persist [2]. Similarly, frequent sprains after the index injury cause chronic ankle instability (CAI) [3, 4]. Symptoms caused by CAI are characterized by the recurrence of pain and swelling during activity and weakening of the muscles around the ankle joint [4].

CAI, which has a high recurrence rate and long-lasting discomfort, can be noted as a characteristic sensorimotor defect. This is a decreased conscious perception, indicating a defect in afferent somatosensory information and efferent motor control [5]. In the study of cortical excitability, there was a difference in cortical excitability and joint laxity between CAI patients and healthy adults, and it was reported that cortical excitability was decreased in ankle instability patients subjected to transcranial magnetic stimulation. This means that the chronic state of joint laxity has high cortical excitability, and it means that cortical excitability as well as joint laxity needs to be reduced [6, 7].

In other words, CAI has great influence on the nervous system, which acts as a modulator of body movement. Therefore, the main rehabilitation protocol for patients with CAI is to improve neuromuscular function. This can be explained by a decrease in joint position sense due to damage to the ankle ligament receptors [8] and peroneal nerve [9] resulting from the injury [10].

Among the conventional interventions for CAI, anterior to posterior talocrural joint mobilization effectively improves CAI and may show some neuromuscular mechanisms by increasing fibularis musculotendinous stiffness [11]. However, the effect of passive joint mobilization (PJM) by the therapist is limited to the neuromuscular mechanism as there is no active movement.

Therefore, active joint mobilization (AJM) is judged by the patient’s active repetition accompanied by pain-free motion to sustained mobilization performed by the therapist and can have a greater influence on cortical excitability. This aims at the central nervous system level of motion corrected under the therapist’s guidance, as re-education on pain-free motion patterns is implied in neuroscience theory [12-15].

In CAI patients, joint mobilization combined with voluntary movement will have more influence on the neuromuscular system than PJM, which corresponds to conventional joint mobilization. Therefore, in this study, the range of motion and balance ability are compared and verified to find out the effect according to with or without voluntary movement.

Study design

This single-blind two-arm parallel design randomized controlled trial included two evaluation sessions (baseline and post-intervention) and 10 intervention sessions (PJM and AJM). In the outcome measure, the primary variable was the dorsi flexion range of motion (DFROM) of the ankle joint, while the secondary variable was dynamic balance ability (DBA).

Participants

Recruitment of potential participants was recruited by staff and students through the bulletin board of Gwangju Health University, Gwangju, Korea. Forty-four potential participants who diagnosed an ankle sprain were recruited; of them, 30 were ultimately enrolled in the study.

The inclusion criteria were: (a) self-reported instability and feeling of “giving way,” (b) a history of sprains in only one ankle, (c) at least two sprains on the same side in the past 2 years, (d) Feeling different in sensation compared to an intact ankle, and (e) not receiving other treatment during the study [16].

The exclusion criteria included: (a) acute ankle sprain within the last 6 months, (b) a history of bilateral ankle injury, (c) bony injury related to ankle sprain, and (d) a history of surgery on the back or lower extremities [16].

Prior to the study, all participants were informed about the purpose and procedure in accordance with the ethical standards of the Declaration of Helsinki. Only those who voluntarily signed an informed consent form were enrolled.

Randomization and blinding

Participants were randomly and blindly assigned to the PJM and AJM groups using random allocation software (ver. 1.0; Isfahan University, Iran).

Intervention

The PJM and AJM interventions were performed on the affected foot for 10 min in each session by two physical therapists who have more than 5 years of clinical experience and have been trained in Orthopedic manual physical therapy. The participants received a total of 10 sessions five times a week for 2 weeks [17].

Passive joint mobilization

PJM uses Maitland’s mobilization method, grade III (high amplitude in the end range of the joint and 1-s vibration in the middle range through linear motion in which tissue resistance is felt). The participant assumed a supine position, and the physical therapist held the talus with one hand and the tibia with the other hand and performed joint mobilization in the posterior direction with the hand holding the talus (Figure 1) [18].

Active joint mobilization

In the AJM, the participant bent their knees in the prone position. The physical therapist held the medial malleolus with one hand and the lateral malleolus with the other. At the same time, the physical therapist put the participant’s soles against the sternum and pressed them in the plantar direction. At this time, the medial malleolus would glide in the anterior direction and the lateral malleolus would glide in the posterior direction. The first procedure passively recognizes participant’s joint motion, while the second procedure involves the voluntary motion(Figure 2) [19-21].

Outcomes

Participants are enrolled in the study and evaluated for DFROM and DBA prior to intervention (baseline), and immediately after two weeks of intervention (postintervention).

Dorsiflexion range of motion

The DFROM of the ankle was measured using a weight-bearinglunge. Thetherapistplacedatape measure on the floor perpendicular toaflat wallwithout protrusion and affixed it toprevent motion.Thesecondtoe andheel of the participant’s foot to be measured were placed parallel to the tape measure, and the non-measured foot was separated from the measuring foot as much as the individual’s foot. The standard for placing the first second toe was 10 cm and bent until the knees touched the wall. The maximum range of ankle instepflexionwas until the heel reached without falling off the floor [22]. As a precaution, the trunk of the body should not be tilted forward and should be bent toward the wall only at the knee.If the knee could reach the wall during bending, the footwas movedforward,while ifthe knee easilytouched the wall,the foot was movedbackward.This measurement was performed three times for each of the leftand rightfeet, and the average valueof the measured valueswas recorded as the ankle angle.

The inter-rater reliability of DFROM has a high intraclasscorrelationcoefficientof 0.99[23], andminimal detectable change is from1.1 to 1.5 cm [24].

Dynamic balance ability

DBAwas evaluated with the participant in a standing position on both feet using the Biodex Balance System (Biodex Balance System; Biodex Medical Systems Inc., USA). The balance measurement equipment consisted of a fixed circular scaffold with a sensor that detects anterior/posterior, left/right motion, a monitor that can visually check and match a target, a computer for data transmission and analysis, and a computer for analysis data output. It consisted of four main types, including printers. The higher the score, the better the balance ability. At the time of the measurement, both arms were collected on the chest, the center of gravity of the body was moved to maintain it according to the target, and the foot position was carefully measured to prevent movement [25].

In the evaluation of the DBA, the scaffold was not fixed, and scaffold instability could be adjusted in various stages from 1 to 12 (least stable to most stable). In this study, DBAwas measured by setting the level of stool movement to one of eight stages. As in the static balance ability evaluation, the foot position was not changed, but the task time and route were recorded by setting the target point and moving the body center to reach the point. Measurements were made twice for 20 s with a 10-s break between them. Measurement of DBA using the Biodex Balance System was reported to have an ICC of 0.42 to 0.80 with appropriate intratester reliability [26].

Sample size

In the study by Cruz-Díaz, Lomas Vega (16), the sample size was calculated by changing the range of motion before and immediately after joint mobilization in CAI patients. The calculated Cohen’s d was 1.36, and when two groups and power were set to 0.95, using G*power 3.1 (G-power 3.1; Heinrich-Heine-Universität Düsseldorf, Germany), 26 samples were required. A total of 30 participants were enrolled in the study according to expected dropout and the central limit theorem.

Statistical analysis

All statistical analyses were performed using IBM SPSS Statistics version 25.0 (SPSS 25.0; IBM Corp., USA). The homogeneity test was performed using the chi-squared test (categorical variables) and independent t-test (continuous variables). An independent t-test was used to verify the effect of the difference between groups, while a paired t-test was used to determine the difference before versus after the interventions. Cohen’s d and partial eta squared (η²_P) were used to determine the effect size of the treatment. η²_Pwas calculated using repeated measures analysis of variance (RM ANOVA). All statistical significance levels (α) were set at 0.05.

This study was conducted between April and June 2019. A total of 30 participants were recruited, and there were no dropouts (Figure 3). Table 1 presents the participants’ general characteristics. There were no significant intergroup differences.

Dorsiflexion range of motion

There was statistically significant improvement in the PJM (d = −1.14, P < 0.001) and AJM (d = −2.90, P <0.001) groups before versus after the intervention. More significant improvement was noted in the AJM group than in the PJM group (η²_P= 0.330, P < 0.01; Table 2).

Dynamic balance ability

In DBA-total, the difference before versus after the intervention was significantly improved in the PJM (d = −1.33, P < 0.001) and AJM (d = −1.61, P < 0.001) groups. There was no significant intergroup difference (η²_P= 0.047, P> 0.05; Table 2).

In DBA-anterior,thePJM group showed no significant improvement (d =−0.71, P>0.05), whereas the AJM group showed significant improvement (d = −2.60, P< 0.001). Therefore, the AJM group showed significantly greater improvement than thePJM group (η²_P =0.434, P< 0.001; Table 2).

In DBA-posterior, significant improvement was seen in both the PJM group (d = −0.74, P < 0.01) and the AJM group (d = −1.84, P < 0.001). The AJM group showed greater improvement than the PJM group (η²_P=0.369, P < 0.001; Table 2).

In DBA-left, significant improvement was seen in both the PJM group (d = −1.55, P< 0.01) and the AJM group (d = −1.73, P < 0.001). There was no significant intergroup difference (η²_P= 0.010, P> 0.05; Table2).

In DBA-right, the PJM group did not show any significant improvement (d = −0.65, P> 0.05), whereas the AJM group showed significant improvement (d = − 2.23, P < 0.001). Significant differences were found before versus after treatment, but a separate sub-analysis (repeated-measures analysis of variance) showed more significant improvement in the AJM group than in the PJM group (F = 8.321, η²_P= 0.229, P < 0.01; Table 2).

In this study, we compared the effects on ankle DFROM and DBA in CAI patients via PJM, a method traditionally performed in previous studies, and AJM, a method that can have a greater effect on neuromuscular system with voluntary movements.

After PJM and AJM were administered a total of 10 times to CAI patients, significant improvement was observed after the intervention in both groups except DBA-anterior and DBA-right in the PJM group. In addition, significantly greater improvement was noted in AJM than in PJM in DFROM, DBA-anterior, DBA-posterior, and DBA-right. Joint mobilization had a large effect on each dependent variable (η²_P > 0.14) [27] of DFROM, DBA-anterior, DBA-posterior, and DBA-right.

Significant improvement in ankle DFROM showed a large effect (d > 0.80) [28] in the PJM and AJM groups. These results are similar to those of previous studies that showed significant improvement when performing anterior to posterior talocrural joint mobilization in CAI were obtained [30, 31]. The minimal detectable change patients [18, 29]. In addition, in the case of voluntary of DFROM for CAI patients reported in previous studies movements, the same resultsas those of previous studies was reportedly0.85 cm[32], which supports the results of this study. This is the basis for the fact that the talus wasglidefromanteriortoposterior withjointmobilization, which is the same method as the previous studies.

The result of the improved ankle DFROMhypothesized that the anterior position fault of the talus changed the arthrokinematic (posterior talar glide) [33]. Therefore, the PJM and AJM involved posterior gliding of the talus and induced significant improvement.

In the DBA variables, the PJM and AJM groups showed significant improvement and had a moderate to largeeffect (0.50 < d< 0.80). This issimilarto the results of Tomruk, Tomruk [34] that ankle joint mobilization with movement in dynamic balance showed significant improvement in postural control. In addition, a number of studies reported significant differences in the star excursion balance test, a variable that can indicate DBA, similar to how AJM showed significant improvement over PJM [16, 35, 36].

In a review article, AJM was effective and widely implemented in musculoskeletal pathologies and dysfunctions of various peripheral joints [12]. Another study reported a difference in the relationship between cortical excitability and joint laxity between healthy adults and those with CAI [7]. This means that the DBA improvement noted in this study affects the movement pattern changed by ankle instability and thus can escape from the chronic state. In a functional near-infrared spectroscopy study, supplementary motor areas (an important structure for motor planning and movement strategies) in CAI patients require activation compared to healthy adults [37].

The main limitation of this study is that itsparticipants were recruited from universities and of similar young age. It was difficult to recruit a large number of participants due to the single-center design, and it may be difficult to generalizeour results to other populations. In future research, hospitals or specific sports groups should also be recruited, and more participants need to be recruited and generalized. And finally, additional brain imaging studies are needed to demonstrate cortical excitability that occurs through voluntary movements.

In conclusion, joint mobilization may be effective at decreasing ankle dorsiflexion and reducing dynamic balance, characteristics of patients with CAI. However, using functional variables such as DBA, it was inferred that joint mobilization using voluntary movements can further stimulate cortical excitability in postural control. Asin the results of this study, when a voluntary movement was made,dynamicbalancewas significantly improved.

The researcher claims no conflicts of interest.