Explanation for the choice of comparators

Participants will be randomized into either the intervention group involving the active PEMF therapy and SC regime or the control group, receiving a sham exposure from the same PEMF device and SC regime.

Intervention description

Participants randomized into the experimental group will be exposed to PEMF therapy employing a BIXEPS device (Quantum TX Pte. Ltd., Singapore) as previously described by Venugobal et al. [19]. Exposure will be given twice a week for 8 weeks for a total of 16 sessions on their affected foot in adjunct with SC. Each treatment session will last for 10 min. The active PEMF therapy does not produce any sensation, facilitating the blinding process. Meanwhile, the control group will receive the SC and the sham treatment from the same PEMF device. A team researcher will operate the PEMF device and monitor any clinical changes during the randomized clinical trial. Each participant will be tagged with a unique radiofrequency identification (RFID) card supplied by Quantum TX Pte. Ltd. using block randomization to assign the treatment group. To mitigate any selection bias, the block size will be concealed from the investigators, and all will be blinded from the subjective assignment.

A team researcher will be responsible for turning on the PEMF machine for all the participants using the RFID cards. A biostatistician not involved in the participant recruitment will oversee the randomization procedure. Both the research team and the biostatistician will remain blinded during the trial. Participants will use the RFID card to complete the treatment without knowledge of group allocation. Side effects will be reported if there are any.

The treatment procedure requires that the participant be seated at 90 degrees on a chair. The solenoids will be adjusted to cover the lower limb. The BIXEPS device is set to low frequency and amplitude fields (1 mT and 50 Hz) to be applied to the affected limb for 10 min. Sham PEMF therapy and SC will be used for the control group. Since the PEMF therapy does not elicit any sensation, both groups are blinded, and the same SC regime will be administered to both groups. Figure 2 depicts how the participants will be situated relative to the BIXEPS device.

Fig. 2

Demonstration on participant being treated with PEMF therapy

The administered SC exercise regime to the participants will be adapted from Alahmari et al. [31] with a detailed regime made into a booklet for the participants. Participants must conduct a warm-up (stretching around the ankle joint) for 5 min before the exercise regime. Participants will be required to engage between the first and 3rd week; participants will be required to participate in balance training while standing on an unstable surface between the first and third week. The treatment regime consists of 5 times/week for 15 to 20 min a day. During the wobble-board training, participants must move from front to back, left to right side for 15 s, 10 repetitions, with 10 s of rest between each repetition. Participants are later required to move in a circle for 60 s, 5 repetitions, with a rest of 20 s between each repetition.

Between the fourth and eighth week, participants will advance to flexing their knees on the wobble board and repeating the same exercise performed during the first to third week. Each repetition will last for 30 s, with a total of 5 repetitions. A rest period of 20 s will be observed between each repetition. Participants must be in a unilateral stance with eyes opened and closed for 7 s and 4 s, respectively, in sets of 5 with a 10-s rest. Concurrent with the balance training, participants will participate in the ankle muscle strengthening exercise using resistant bands (Thera-bands). Participants are required to sit on the floor with one end of the tubing tied around the treatment table and the other wrapped around the affected foot’s involved metatarsals. Participants will perform different ankle rotations using resistant bands, such as eversion, inversion, plantarflexion, and dorsiflexion. Regardless of the tension of the band, participants are required to stretch the band at 170% of the resting length. Exercise progression will include increasing the number of sets from 1 to 3 for 8 to 10 repetitions each—lastly, stretching of the ankle joint during the cooling session. The major exercise regime can be seen in Fig. 3.

Fig. 3

Resistance training and balance training

Criteria for discontinuing or modifying allocated interventions

A participant can opt out of the study at any time point. Treatment will be halted immediately, and required treatment will be provided whenever there is an adverse reaction, including pain, swelling, and discomfort. Consultations with the orthopedic surgeons will be made available via the research clinic from the Department of Orthopaedics and Traumatology, CUHK.

Strategies to improve adherence to interventions

The opening hours of the Sports Performance and Biomechanics Laboratory will be extended to evening hours for the convenience of participants and to facilitate their participation. In addition, participants will be asked if there is any difficulty with the exercise regime whenever they come for the PEMF therapy.

Relevant concomitant care permitted or prohibited during the trial

Participants are allowed to participate in their usual activities; no restrictions or changes to their daily activities will be required.

Provisions for post-trial care

Participants will have access to the research clinic from the Department of Orthopaedics and Traumatology, CUHK, if they experience any adverse reaction.

Outcomes Primary outcome

Single leg standing test will be assessed on the TekScan MatScan® pressure mat model 3150 (TekScan Inc, South Boston, USA) to evaluate the postural control stability. The TekScan MatScan® is a low-profile floor mat (5 mm thick) and consists of 2288 resistive sensors (1.4 sensors/cm2) with a sampling frequency of 40 Hz and has been well-validated in other populations [32]. The mat will provide measures of anterior–posterior and mediolateral sway parameters described as the area and direction of the sway, distance and the direction traveled by the center of pressure (COP), and the variability of the distance traveled by COP. The Sway Analysis Module (SAM™) software will be used to analyze the sway data as seen in Fig. 4.

Fig. 4

A TekScan result for anterior–posterior and mediolateral direction

On the pressure mat, participants will perform one practice trial and two trials of single limb stance on each leg with eyes opened and closed for 20 s, respectively. Participants will be required to stand on one leg unassisted (barefooted) and hands on the hips. This will be repeated for the non-affected leg as well. Before the commencement of the study, the assessor will undergo training, and the same assessor will oversee the experiment.

Secondary outcomes Functional score

Cumberland Ankle Instability Tool (CAIT) score is a valid and reliable 9-item instrument used to identify self-perceived ankle instability. The instrument is scored on a 0 (worst) to 30 (best) scale, where a cut-off point of ≤ 24 denotes the presence of instability [33].

Dynamic balance

The Y balance test (YBT) evaluates the dynamic balance control in the CAI population. This test requires the integration of sensory inputs from visual, vestibular, and somatosensory pathways resulting in controlled motor outputs that recruit specific motor inputs. Participants must maintain unilateral stance, with their hands on the hips, balance on the involved limb, and reach with the uninvolved limb in the anterior, posteromedial, and posterolateral directions as far as possible [34]. The trial will not be counted and repeated if the participants lift their heels, fail to maintain their balance, remove their hands from their hips, or return to the starting point. Participants will complete 4 times of the trial and 3 analysis trials in each direction of the limb. The maximum distance in each direction, mainly, anterior, posteromedial, and posterolateral will be taken and divided by the limb length (LL) and multiplied by 100 [35]. LL will be measured using a flexible measuring tape from the anterior superior iliac spine to the medial ankle. The YBT composite (%) is calculated by the sum of three reach directions divided by 3 times the LL multiplied by 100.

Eversion muscle strength

Peroneal longus (PL) contribute greatly to ankle eversion; ankle eversion muscle strength will be measured using an isometric digital handheld dynamometer (MicroFET2, Manual Muscle Testing) under a standardized protocol [36, 37]. Participants will be placed in a prone position, and the ankle posture will be anatomically neutral. A handheld dynamometer will be placed on the lateral metatarsal head for the eversion testing. The patient will exert muscle strength, and the examiner will counteract with resistance for 5 s and measure the muscle strength isometrically. The healthy ankle will be examined first, followed by the injured ankle. The test will be performed thrice, with the highest value recorded in kilograms. The maximum strength (kg) will be taken as the participant’s peak strength and the result will be normalized by the body weight (kg). Both the injured and contralateral limbs will be tested.

Dorsiflexion range of motion

Weight-bearing lung test is based on the knee-to-wall principle and is a valid and reliable assessment that measures the DROM in CAI population. The procedure requires the participant to first perform a lunge in for the untested limb where the knee flexes to a point for the anterior knee to touch the wall with the tested heel remaining firmly on the floor. The maximum distance is determined from the tip of the great toe to the wall, nearest to 0.1 cm, using a long ruler [38]. Upon successful contact of the knee to the wall, the participant will progress backward in 1 cm increments from the wall until either the heel or the knee can no longer be maintained during the lunge. The participant will be given 3 attempts to achieve the greatest distance between the large toe and the wall. Both the injured and contralateral limb will be tested.

Lateral step-down test (LSDT)

LSDT is a clinical measurement that evaluates dynamic postural stability deficits and potential lower extremity injury. It measures the coordination of the lower limb joints, neuromuscular control, strength, pelvic stability, and range of motion [39]. A 30-cm high step will be used for the test, and the participant will be instructed to align the stance limb foot with the lateral edge of the step with their hands on their hips. Participants will also be instructed to descend the step with the tested heel gently touching the TekScan MatScan® pressure mat model 3150 (TekScan Inc, South Boston, USA), with the contralateral foot lifted off the platform. The trial will not be considered successful when the hands were not placed at the iliac crest during the starting point, the testing leg touches the pressure mat before the opposite leg left the step, or the participant jumped instead of stepping down. Participants are required to balance on the contralateral heel for 20 s, and the COP will be captured by the Sway Analysis Module (SAM™) software. Both the injured and contralateral limb will be tested; participants will repeat the test 3 times on each leg.

Ultrasonographic outcomes

A standardized ultrasound examination is used to evaluate the peroneal longus (PL) in a reliable and repeatable manner. It will be performed during baseline assessment and all follow-up visits. The overall ultrasound examination has shown high intra-reliability (ICC = 0.925, p < 0.01) and inter-rater reliability (0.947, p < 0.01) and was conducted together with a researcher (HX) who has 5 years of experience with ultrasound imaging.

Cross-sectional and thickness of the peroneal muscle

Musculoskeletal ultrasound scanning will be performed in the supine position with the probe at 25% of the distance between the fibula head and lateral malleolus. Peroneus longus (PL) and peroneus brevis (PB) were scanned together transversely by positioning the long axis of the probe perpendicular to muscle fibers to obtain the image of the cross-sectional area (CSA) [40]. CSA will be taken as the area of muscle perpendicular to its longitudinal direction along the line between the fibular head and lateral malleoli. The probe will be rotated into a longitudinal orientation for the thickness of the peroneal longus (PL). The thickness will be defined as the distance between each peroneal muscle’s aponeuroses (white fibrous tissue) [40].

The elasticity of the peroneal muscle

In addition, the shear elastic modulus of a muscle will be measured using ultrasonic shear wave elastography (Aixplorer, Supersonic Imagine, France) with a linear array probe. The region of interest (ROI) will be set near the center of each muscle. The analysis area will be a 5-mm diameter circle at the center of ROI. The shear elastic modulus will be measured three times per muscle, and the average of the three measured values will be used for further analysis. The shear elastic moduli of the PL and PB will be measured at proximal 25% from the head of the fibula to the lateral malleolus. For each Q-box, the maximum, minimum, and mean (in kPa) of the stiffness of the muscle are measured [41].

Other patient-reported outcome measures (PROM) will be employed to assess overall health and foot and ankle-related symptoms, pain, physical activity level, and health-related quality of life. Foot and Ankle Ability Measure (FAAM) will be used to assess the foot and ankle functional limitations among CAI individuals. The FAAM consists of 29 items, scored between 0 and 4, divided into two subscales: activities of daily living (21 items) and sports (8 items) [42]. For score analysis, the percentage of each subscale is used separately. According to the International Ankle Consortium, values of the day living activities < 90% and < 80% of the sports subscales of the FAAM score were considered unsatisfactory. Manchester-Oxford foot and ankle questionnaire (MOXFQ) is a 16-question self-reported score designed to evaluate the impact of foot and ankle problems on a person’s quality of life. The dimensions comprise 3 subscales: walking/standing problem (7 items), foot pain (5 items), and issues related to social interaction (4 items). After that, all three domains of the MOXFQ can be summed up and converted to a metric from 0 to 100 to create a summary index score (MOFXQ index) [43]. The pain score will be evaluated using the visual analog scale (VAS) ranging from 0 to 100. For the pain score, the higher the score indicates, the more severe the pain. The Short Form-36 questionnaires (SF-36) will be used to evaluate the health-related quality of life [44]. The International Physical Activity Questionnaire will be used to evaluate the physical activity levels [45]. All primary and secondary outcomes will be evaluated at baseline, 4th week, 8th week, 3rd month, 6th month, and 12th month follow-up.

Participant timeline

The schedule of study outcomes is assigned in Table 1.

Table 1 The enrolment, intervention, assessment, and visit schedule for the participantsSample size

Using the G*Power software (Germany) [46], a sample size of 48 (n = 24 per group, 1:1 allocation) is calculated based on the highest attrition rate of 40% at 3rd month follow-up of the study to retain a sample size of n = 28, according to similar research [47]. The calculation is according to the F test-repeated measures ANOVA, between factors, with an effect size of 0.3, an alpha level of 0.05, (1 − β err prob) = 0.8 to detect a significant change of the area of COP from 42.0 (± 23.6) to 25.4 (± 8.2) [48].

Recruitment

Participants will be principally recruited from the Department of Orthopaedics and Traumatology, Prince of Wales Specialist Outpatient Clinic. The orthopedic surgeons will screen the participants based on their clinical symptoms and explain the details of the study to the participants. Furthermore, the recruitment information for the trial will be delivered to various sports teams and through the internal mass-mail system at CUHK. They will be referred to the research team for eligibility assessment at the Sports Injury and Biomechanics Laboratory should they want to join the trial.

Assignment of interventions: allocationSequence generation

Participants will be randomly assigned to a unique radiofrequency identification (RFID) card number in a 1:1 ratio by an online research randomizer (https://www.randomizer.org/). The unique RFID card is provided by the Quantum TX. Each allocation will be tagged with a unique RFID card generated by the chief engineer from Quantum TX and is recognized by the PEMF machine.

Concealment mechanism

The randomization will be conducted using an online research randomizer by the chief engineer at Quantum TX and the random order of the treatment allocation will only be shared with the biostatistician, a third party based at the CUHK who is not involved in the recruitment process. The randomization will not be disclosed to either the participants or the trial assessors throughout the intervention period. The participants will be randomly assigned to a unique and identical RFID card in which the PEMF or the sham group are randomly allocated to the RFID card. This is to ensure that the participants and the assessors are both blinded to the unique RFID cards containing the treatment types.

Implementation

The chief engineer of Quantum TX will generate the randomization, and the randomization results will only be shared with the biostatistician at the CUHK. The research team and the orthopedic surgeons will recruit the participants. The Ph.D. student will assist in the assessment and assign the participants to the interventions. At the end of the intervention, the biostatistician will reveal the randomization results to the research team and orthopedic surgeons for statistical analysis.

Assignment of interventions: blindingWho will be blinded

The assessors and orthopedic surgeons will be blinded to the assignment of the interventions. Since the PEMF therapy will not elicit heat or sensation, the participants will also be blinded to the treatment allocation. The randomization results will only be disclosed to the assessors and the orthopedic surgeons after the PEMF intervention has ended.

Procedure for unblinding if needed

The procedure will only be unblinded if there is an occurrence of an adverse event. The adverse event will be reported to the Joint Clinical Research Ethical Committee of the CUHK and the New Territories East Cluster of the Hospital Authority.

Data collection and managementPlans for assessment and collection of outcomes

Demographic data and injury history will be collected during screening and before the recruitment. Participants who have fulfilled the eligibility criteria will be randomly allocated into the treatment groups. This is followed by the collection of the outcomes as listed in “Results” at baseline and different time points.

Plans to promote participant retention and complete follow-up

The office hours of the Sports Performance and Biomechanics Laboratory, Prince of Wales Hospital, will be extended to the evening to accommodate the participants’ working hours for the PEMF therapy and assessments.

Data management

The collected data will be entered into Excel and SPSS files. They will all be password-protected and destroyed 5 years after publication in Ph.D. thesis and peer-reviewed journals. The Ph.D. student will be responsible for the data collection, data entry, and data analysis under the guidance of the biostatistician. Double data entry and range checks for data values will promote data quality.

Confidentiality

All online data relating to the participants will be password-protected on the laboratory’s computer, and hard copies (i.e., consent forms and data sheets) will be placed in files and kept in a locked cabinet at the Sports Performance and Biomechanics Laboratory. Only the research team in charge of the study will have access to the information. Participants will not have any access to the data set. There is no plan to share the data collected with other organizations.

Plans for collection, laboratory evaluation, and storage of biological specimens for genetic or molecular analysis in this trial/future use

There are no plans to collect any biological specimens for genetic or molecular analysis in this trial.

Statistical methodsStatistical methods for primary and secondary outcomes

SPSS version 26.0 will be used to analyze the primary and secondary outcomes based on the intention to treat (ITT) principle. ITT analysis will be performed using all randomized patients, including data for those terminated or lost to follow-up. During the process, assumptions will be checked, and remedial measures will be employed. The normality of the data will first be determined using the Kolmogorov–Smirnov test or the Shapiro–Wilk test (p > 0.05) to test for the normality of the data distribution. Other graphical methods such as Q-Q plots or histograms will also be used for visual inspection of the distribution and to assess the normality. If skewness and heteroscedasticity are present, transformation methodology such as Logarithm may be applied to make the distribution more symmetric and stabilize the variance. Baseline data and clinical variables between both groups will be compared using independent Student’s t-tests (parametric)/Mann–Whitney U test (non-parametric) for continuous data and the χ2 of independence for categorical data, respectively. Descriptive statistics will be presented in either mean (SD) or frequency (%). Thereafter, a repeated measure analysis of variance (RMANOVA) will be used to measure the time/exposure effects of muscle strength, Y balance test, LSDT, SLT, DROM, and results of the questionnaire between baseline, week 4, week 8, 3rd month, 6th month, and 1-year follow-up. Analysis of covariance (ANCOVA) will be conducted for baseline characteristics and prognostic factors as covariates and compared for group differences in different time points. Repeated-measure ANOVA and ANCOVA will be applied for all participants who have completed the assessments post-treatment. Bonferroni correction will be performed for multiple comparisons. Sensitivity analysis will be conducted by comparing the results of the imputation method to complete-case and available data analyses. The significant level will be set at 0.05.

Interim analyses

Interim results will be performed when 20% of the participants have completed the 8-week PEMF treatment. The interim analysis ensures that the study is feasible and safe during the study period.

Methods for additional analyses (e.g., subgroup analyses)

No subgroup analysis will be performed.

Methods in analysis to handle protocol non-adherence and any statistical methods to handle missing data

Intention to treat analysis will be deployed to include all study participants regardless of their treatment allocation and whether the intervention for them has been completed. Regardless of the adherence to the intervention or withdrawal, they will be recorded by the Ph.D. student weekly. Missing data will be checked, and multiple imputations and sensitivity analysis will be employed assuming that the missing data are random.

Plans to give access to the full protocol, participant-level data, and statistical code

The protocol has been registered on the ClinicalTrials.gov (ID: NCT05500885). The study data will be available from the principal investigator upon reasonable request.