The effects of isolated ankle strengthening and functional balance training on strength, running mechanics, postural control and injury prevention in novice runners: design of a randomized controlled trial

Muscle strength

It has been found that weak plantar flexor strength is associated with an increased incidence of Achilles tendinopathy, one of the most common overuse running injuries[16]. Furthermore, decreased muscular strength, particularly at the quadriceps and hamstrings, has been associated with increased incidence of patellofemoral pain syndrome, another common overuse running injury[17]. Reduced hip abduction strength has also been found in subjects suffering from patellofemoral pain syndrome and has generally been found to be a strong predictor of lower extremity injury risk[19, 20].

Joint kinematics and kinetics

Research examining joint kinematic and kinetic variables during running has found that increased hip adduction and knee internal rotation are risk factors for the development of iliotibial band syndrome in runners[18]. Furthermore, it is thought that abnormal movement of the tibia and femur in the frontal and transverse planes affect patellar tracking and, therefore, increase the risk of suffering from patellofemoral pain syndrome[21]. This has been seen in runners with patellofemoral pain syndrome who demonstrated increased ankle eversion velocities and altered joint coupling mechanics[22]. It also has been found that runners with patellofemoral pain demonstrate increased knee joint angular impulse in the frontal plane compared to healthy runners[23].

Most of the recent running injury prevention studies have focused on hip strengthening protocols with the hope of reducing excessive joint movements and moments[24, 25]. The strengthening of the hip joint musculature assumes a “top down” approach where it is thought that improving strength at the hip will not only reduce movements and loading at the hip, but also at the more distal knee and ankle joints[26–28]. However, hip strengthening protocol interventions demonstrated that an increase in hip muscular strength was not accompanied by a reduction in excessive joint movements or moments in the frontal and transverse planes during running[24]. One potential explanation for these findings would be that the strengthening protocols were focusing on the proximal hip joint. It may be that a “ground up” approach should be used, in which specific ankle strengthening would be the primary focus of training. Such an approach would hypothesize that changes at the ankle joint will influence the mechanics of the more proximal knee and hip joints[27, 29]. Biomechanical injury prevention models focus on adjusting both the external and internal loads on the body[30]. In this sense, interventions aim to reduce the loads below the injury threshold or increase the body’s ability to tolerate the load[30]. It has been proposed that isolated ankle strengthening increases the strength of the small muscles surrounding the ankle joints, which are more adept for making quick postural adjustments without excessive force[31]. The use of these small stabilizing muscles rather than the larger muscles with longer lever arms may increase joint stability and reduce the loading at the joint and insertions of the larger muscles surrounding the ankle joint[31]. As a result, increased ankle musculature strength may not only increase the body’s ability to tolerate the load through an increased muscle ultimate strength but may also reduce the internal loads at the joint and at the muscle insertions below the injury threshold[31].

Another potential explanation for the lack of kinematic and kinetic changes found with hip musculature strengthening may be that the strengthening protocols used were not functional for an activity such as running. The principle of specificity dictates that training protocols should mimic the movements of the particular sport as closely as possible in order for the body to adapt appropriately[32]. Clinicians accept the importance of functional training and incorporate functional movements into training protocols, however there is a paucity of literature evaluating such functional training programs[33, 34]. Some hip strengthening protocols primarily focus on open chain, non-weight bearing exercises to strengthen the hip and thigh musculature[24]. Running is a closed chain, weight-bearing exercise and, therefore, training that does not incorporate similar mechanics that replicate the running movement may not alter running kinematics and kinetics. The only closed chain, weight-bearing exercise used by Willy and Davis was the single leg squat movement re-education protocol. The only movement differences found were those during the single-leg squat rather than during running. Similar results have been found previously indicating that isolated strengthening programs may not affect movement mechanics[35]. In order to alter lower extremity movement mechanics, training protocols may need to incorporate functional movements in order to improve neuromuscular control at the joint.

One form of training that may be able to accomplish this functional requirement is the incorporation of an unstable device into the training protocol. The appropriate training program using an unstable device should be able to incorporate functional closed chain, weight-bearing exercises that replicate the running movement to possibly increase strength and reduce excessive joint movements and moments during running. Functional balance training has traditionally been used in ankle injury rehabilitation programs with patients suffering from chronic ankle instability and repeated ankle sprains[36, 37]. One main goal of rehabilitation programs was to increase the strength of the ankle joint following injury. Research examining the effects of such training on lower extremity strength has shown successful increases in muscle strength[38, 39]. In recent years, functional balance training protocols have been utilized for the prevention of lower extremity injury in sport for healthy, physically active participants. Multifaceted training programs have seen positive results with respect to the reduction of excessive joint movements and moments as well as the incidence of acute and overuse lower extremity injuries in sport[40–42]. The same group that had utilized a multifaceted training program to reduce knee valgus in high school athletes went on to investigate the effects of the balance training portion of their protocol on knee valgus during landing[43]. For the same duration and intensity of training that was used in the multifaceted approach, it was found that balance training alone reduced hip adduction and ankle eversion angles during landing from a drop vertical jump. Similar results have been found in other functional balance training programs where frontal and transverse plane movements and joint moments have successfully been reduced. Even more importantly, functional balance training has successfully reduced the incidence of acute and overuse injuries in sports such as soccer, volleyball and basketball where non-contact ACL injuries and patellofemoral pain syndrome are common[42, 44–50]. However, such an intervention has yet to be evaluated in a recreational running setting.

Study design

This is a randomized controlled trial (RCT). In total, 120 novice runners will be recruited. Subjects will be randomly assigned to one of three groups: an isolated ankle strengthening group (n = 40), a functional balance training group (n = 40) or an activity matched control group (n = 40). The study design and flow can be seen in Figure 1. Research ethics approval was obtained from the University of Calgary’s Conjoint Health Research Ethics Board (Ethics ID: REB13-0153).

Target population

The target population for this research project is novice recreational runners. Novice runners have been chosen for this study as they may be the most malleable for training effects. They have also been shown to have a higher incidence or running-related injuries than experienced runners[51–54]. A novice recreational runner will be identified as an individual with less than one year of running experience that runs at least twice a week for a total weekly distance greater than 5 kilometers as their main physical activity[54]. An age range of 18 – 60 years of age will be used for subject inclusion criteria. Inclusion criteria include the lack of any neurological abnormalities or any lower extremity injury or pain within the three months prior to testing and subjects should not have extensive prior experience with isolated joint strength training functional balance training within the last year.

Sample size

Exploratory power analyses were completed for multiple biomechanical variables that will be tested in this study. Sample size was determined based on a sample size calculation using previously published data on the effects of balance training on the 95% ellipse area of the center of pressure as this was the largest reasonable sample size needed for the variables in this study[37]. It was found that in order to achieve 80% statistical power with an α = 0.05, 32 subjects would be needed to detect a change between training group and the control group with respect to the change in the 95% ellipse area following training with an effect size of 0.71. Assuming a 20% drop out rate, 40 subjects will be recruited for each of the groups in this project. Due to the sample size, the injury portion of this study will be a pilot RCT for developing a running injury prevention program. The information found in this study regarding the incidence of injury in each group will be critical for developing future RCTs investigating the impact of these interventions on running injury prevention.

Randomization procedure

In order to protect against selection bias, subjects will be randomly assigned to one of the three groups. The generation of an unpredictable randomized allocation sequence will be completed using randomization.com through the use of a random allocation scheme (one block, size 120) with a 1:1:1 allocation ratio between the three groups. This randomization procedure was chosen to avoid selection bias with predictable random assignments using smaller block sizes in a non-blinded study design. This task will be completed by an individual that is not involved in the trial. In order to ensure allocation concealment, sequentially numbered, opaque, sealed envelopes (SNOSE) will be used. Once potential subjects have provided written, informed consent, they will sequentially draw an envelope and write their details on that envelope before opening to determine their group allocation. All envelopes will be similar in appearance and weight and cardboard will be placed within the envelope in order to make it impermeable to light.

Treatment arms

A home-based warm up routine consisting of aerobic activity, static stretching and dynamic stretching will be taught with recommended adherence by all three groups five times a week[41] The control group subjects will be asked to complete this warm up for twenty five minutes with no additional training (Table 1). The strength group and the balance group will be asked to complete specific training program components for 20 minutes in addition to a 5-minute warm up. The strength group will complete a training program using Thera-Band elastic bands[55] (Figure 2). The balance group will have training incorporating a Both Sides Up (BOSU) device with activities increasing in difficulty every two weeks[43, 46, 56] (Table 2). Subjects will meet in person with the study coordinator every two weeks to receive their new training exercises. All subjects will be asked to complete their respective routines five sessions per week. If the subject plans to run that day, they will be asked to complete their training immediately before their running session. If the subject does not plan to run that day, they are able to complete their respective training whatever time of day is convenient. Subjects will be asked to complete a physical activity log documenting the number of minutes completed for each exercise in their training routine as well as the number of minutes of running exposure on a weekly basis[42]. It is hypothesized that the common warm up will have no effect on muscular strength, lower extremity running mechanics, postural control, or injury prevention.

Category	Movement	Duration
Aerobic	Side-to-side shuttle, high knee skipping, light running	5 min
Static stretch	groin, hamstrings, quadriceps, calves (30 seconds each)	10 min
Dynamic stretch	Buttock Kicks – 4 minutes	10 min
	Leg Swings – Side, 3 minutes
	Leg Swings – Front, 3 minutes
	Total Time	25 min

Isokinetic strength

As running is dynamic, isokinetic measurements of strength will be completed. Isokinetic strength will be measured using a Biodex System 3 dynamometer (Biodex Medical System Inc, New York, NY, USA) to assess average peak torque for ankle eversion/inversion strength, ankle plantar flexion/dorsiflexion strength, knee flexion/extension strength and hip ab/adduction strength. For each movement, three maximal repetitions of concentric-concentric contractions will be completed. Three submaximal contractions will be allowed prior to three maximal contractions to allow for familiarization with the procedure. Isokinetic dynamometer results have previously been shown to produce reliable measurements for isokinetic strength[60, 61]. For ankle testing measurements, subjects will be placed in a semi recumbent position with 60° of knee flexion. Strength measurements will be made at 30°/s. For knee flexion/extension isokinetic strength, subjects will be placed in an upright seated position with knees and hips flexed to 90°[62]. Strength measurements will be made at 60°/s. For hip ab/adduction isokinetic strength, subjects will be standing in a vertical position and measurements will be made at 30°/s. Consistent verbal encouragement will be given in order to promote maximal effort.

Postural control

To assess postural control, subjects will be asked to complete three 70-second single limb stance trials with their eyes open and three 45-second single limb stance trials with their eyes closed standing on a force platform embedded in the laboratory floor (Kistler Instruments AG, Winterthur, Switzerland). Only the dominant limb will be used as assessed using the following three tasks (2 out of 3): preferred kicking limb, step up test, and balance recovery[63]. The magnitude of movement and the temporal structure of the center of pressure (COP) trace will be analyzed. The magnitude of movement will be quantified using the path length and 95% ellipse area. The temporal structure will be quantified using the entropic half-life [E(1/2)] of the COP data in the medio-lateral and anterior-posterior direction[64, 65]. Additionally, postural control will be assessed using the Star Excursion Balance Test (SEBT), which has previously been shown to be a reliable and valid measure of postural control[66, 67].

Kinematics and kinetics

To assess lower extremity running mechanics, subjects will perform 20 over-ground running trials (3.5 m/s ± 15%). Additionally, subjects will perform five minutes of treadmill running at their preferred speed following a five minute walking warm up at 2.5 mph. For the over ground running trials, kinematic data will be collected using an eight high-speed video camera system (Motion Analysis Corporation, Santa Rosa, CA, USA) at a sampling rate of 240 Hz. Previous studies have found that motion capture systems provide results with clinically acceptable errors and relatively high reliability depending on the accuracy of marker placement[68]. Three-dimensional marker traces will be reconstructed using Expert Vision Three-Dimensional Analysis software (Motion Analysis Corporation, Santa Rosa, CA, USA). Kinetic data will be collected simultaneously using a force plate (Kistler Instruments AG, Winterthur, Switzerland) embedded within the laboratory floor with a sampling frequency of 2400 Hz. Retro-reflective markers will be mounted on the rearfoot, shank, and thigh of the right lower extremity and the pelvis to measure three dimensional movements of the ankle, knee, and hip joints[69]. Markers will be placed over the right greater trochanter, medial and lateral knee joint axis, and medial and lateral malleoli in order to define joint centers. Position data from a static neutral trial will be collected in order to define the segment coordinate system. Joint markers will be removed for the running trials. For the treadmill walking, 26 retro-reflective markers were skin-mounted the arms, torso, pelvis, legs and feet. Markers were placed on both sides of the body (three on each foot, three on each shank, three on each thigh, one on each greater trochanter and lateral knee joint center, four on the pelvis) as well as two on the sternum, one on the spine (C7), and one on each shoulder. All biomechanical testing sessions will be completed by the same researcher in order to ensure maximum reliability.

Injury assessment

Participants will be asked to complete their allocated treatment two days per week before their running sessions for maintenance following the initial 8 weeks of training. The primary outcome measure for this portion of the research project will be running-related injuries. During the 6-month follow up period, running injuries will be tracked and documented for each participant using StudyTRAX, a web-based Electronic Data Capture system. Injuries will be identified using the “any physical complaint” definition as “any physical complaint developed in relation to running activities and causing restriction in running distance, speed, duration, or frequency”[70, 71]. Injuries reports will be completed weekly using the questionnaire that has previously been validated for overuse injuries[72]. An additional question will be added to this questionnaire asking the participant if they experienced any “physical complaint that prevented them from being fully able to participate in running activities”[72, 73]. If a participant answers “yes” to the final question, they will be contacted directly and asked to fill out the “time-loss” injury questionnaire using a “time-loss injury” questionnaire that was adapted for running injuries[74]. All “time-loss” injuries occurring during this time will be assessed by a physiotherapist. Injury incidence and severity for overuse injuries will be quantified by the average weekly incidence and severity scores from the overuse injury questionnaire[72]. Subjects will be asked to complete a physical activity log documenting their running routines as the number of minutes of running exposure on a weekly basis[42]. This will provide information regarding the running frequency per week as well as the length of time of each run in minutes. Due to funding limitations, heart rate monitors and GPS tracking were not available. Therefore, no information will be available regarding the type of run (tempo, steady state, intervals). Injury incidence and severity for acute injuries will be quantified by injury rate (number of injuries/1000 running hours) and the number of days taken away from running due to the injury[72]. Participants will also self-report their protocol adherence every week during the eight week intervention[75].