Midfoot joint impairment is likely following lateral ankle sprain (LAS) that may benefit from mobilization.
To investigate the effects of midfoot joint mobilizations and a one-week home exercise program (HEP) compared to a sham intervention and HEP on pain, patient-reported outcomes (PROs), ankle-foot joint mobility, and neuromotor function in young adults with recent LAS.
All participants were instructed in a stretching, strengthening, and balance HEP and were randomized a priori to receive midfoot joint mobilizations (forefoot supination, cuboid glide and plantar 1st tarsometatarsal) or a sham laying-of-hands. Changes in pain, physical, psychological, and functional PROs, foot morphology, joint mobility, pain-to-palpation, neuromotor function, and dynamic balance were assessed pre-to-post treatment and one-week following. Participants crossed-over following a one-week washout to receive the alternate treatment and were assessed pre-to-post treatment and one-week following. ANOVAs, t -tests, proportions, and 95% confidence intervals (CI) were calculated to assess changes in outcomes. Cohen’s d and 95% CI compared treatment effects at each time-point.
Midfoot joint mobilization had greater effects (p < .05) in reducing pain 1-week post ( d = 0.8), and increasing Single Assessment Numeric Evaluation (immediate: d = 0.6) and Global Rating of Change (immediate: d = 0.6) compared to a sham treatment and HEP.
Midfoot joint mobilizations and HEP yielded greater pain reduction and perceived improvement compared to sham and is recommended in a comprehensive rehabilitation program following LAS.
A trial was performed assessing midfoot joint mobilization after ankle sprain vs. sham.
Midfoot joint mobilizations yielded greater pain reduction and perceived improvement.
Midfoot joint mobilizations are recommended with exercises following ankle sprain.
Lateral ankle sprains (LAS) are common musculoskeletal injuries incurred by athletes ( ; ) and the general public ( ). LAS involve high-velocities and extremes of rearfoot inversion and tibial external rotation ( ; ) that result in injury of the lateral talocrural ( ; ) and midfoot ( ; ; ) ligaments. Forty-percent of LAS will progress to chronic ankle instability (CAI) ( ), a clinical condition described by perceived or episodic ankle giving-way, persistent activity limitation, and participation restriction that persist beyond one-year post injury ( ).
Midfoot injury is common following LAS and may contribute to patient signs and symptoms ( ). Of patients who incurred LAS, 21–41% had midfoot ligamentous involvement ( ; ) and 33% had midfoot joint capsular injury ( ). Approximately one-quarter of individuals with isolated calcaneocuboid ligament injury were initially diagnosed as having a LAS ( ). Midfoot injury may contribute to or mimic LAS symptoms ( ). Similar to LAS, persistent pain, swelling, giving-way, and repeat episodes of injury are potential consequences following midfoot injury ( ). Cuboid Syndrome, a disrupted congruency of the calcaneocuboid joint, is another potential consequence following LAS ( ). This syndrome is theorized to result from rearfoot inversion with the forefoot loaded (relative forefoot eversion) ( ), or an eversion moment of the cuboid during a forceful fibularis longus contraction during LAS ( ). Joint mobility assessment of all segments of the ankle-foot complex, especially the midfoot and forefoot, has been recommended for all LAS patients regardless of symptom presentation ( , ).
In a recent cross-sectional study of ankle-foot morphology and function comparing young adults with and without a history of LAS and CAI, the LAS group with a 2–8 week chronicity demonstrated increased forefoot eversion and tarsometatarsal motion and no differences in total joint excursion compared to healthy controls ( ). Diminished forefoot inversion and 1st tarsometarsal plantar joint play measures were also observed ( ). These findings were postulated to result from a shift in relative motion further into eversion ( ), consistent with the mechanism thought to cause Cuboid Syndrome ( ).
Joint mobilization, stretching, and strengthening exercises have been suggested for patients with midfoot joint impairment ( , ; ). However, evidence for midfoot joint mobilization and exercise following LAS is limited ( ; ). Therefore, the purpose of this crossover clinical trial was to investigate the effects of midfoot joint mobilization and a one-week home exercise program (HEP) compared to a sham intervention and HEP on patient-reported and clinical measures in young adults with recent history of LAS.
A laboratory-based, crossover clinical trial was performed where the independent variable was treatment (50% allocated to initially receive joint mobilization, 50% allocated to initially receive sham). The primary dependent variables were changes in patient-reported pain and function, foot morphology (foot mobility magnitude, arch height flexibility), joint motion (weight-bearing dorsiflexion, rearfoot goniometry, forefoot inclinometry, 1st metatarsal displacement), strength (handheld dynamometry), and dynamic balance (Star Excursion Balance Test, SEBT) immediately post-treatment and one-week following. Crossover design was selected over a parallel randomized control trial to ensure the individual factors of joint phenotype, injury heterogeneity, and psychological factors ( ) were accounted for in the design. Controlling these factors superseded the risk of carryover effects or the natural healing that may have occurred during the two-week study period. The trial was registered with the National Institutes of Health (NCT02697461). The study was approved by the university’s Institutional Review Board.
A convenience sample of 17 recreationally-active individuals (8 males, 9 females) aged 18–35 with a recent history of LAS were recruited at a public university. Recreationally-active was defined as participation in some form of physical activity for at least 20-minutes per day, at least three times a week. Participants who sustained LAS in the past 2–8 weeks and presented with forefoot-on-rearfoot hypomobility ( ) were included. Participant demographics and self-report measures are in Table 1 . Individuals were excluded if they had a self-reported history of leg or foot fracture, neurological or vestibular impairment that affected balance, diabetes mellitus, lumbosacral radiculopathy, soft tissue disorders such as Marfan or Ehlers-Danlos syndrome, any absolute contraindication to manual therapy, or if they were pregnant. Participants who met inclusion criteria provided informed consent. Fig. 1 details the CONSORT ( ) flow chart from recruitment to analysis.
|Participants allocated to receive sham intervention first (n = 8) |
4 males 4 females
|Participants allocated to receive midfoot mobilization first (n = 9) |
4 males 5 females
|p||Total Sample (n = 17) |
8 males 9 females
|Mean ± SD||Mean ± SD|
|Age (years)||20.4 ± 1.3||23.2 ± 5.3||.17||21.0 ± 2.3|
|Height (cm)||170.2 ± 8.8||174.1 ± 9.5||.39||172.3 ± 2.2|
|Weight (kg)||67.1 ± 13.5||75.7 ± 10.6||.17||71.6 ± 12.5|
|BMI (kg/m 2 )||23.1 ± 3.4||25.0 ± 3.9||.29||24.1 ± 3.7|
|Foot Posture Index||2.0 ± 3.6||3.0 ± 3.2||.56||2.5 ± 3.4|
|Ankle sprains (n)||2.5 ± 1.4||3.2 ± 3.0||.54||2.9 ± 2.3|
|Time to injury (months)||0.9 ± 0.5||0.9 ± 0.7||.84||0.9 ± 0.6|
|IdFAI||26.6 ± 3.5||20.7 ± 4.6||.009||23.9 ± 4.7|
|Kinesiophobia (TSK-11)||23.9 ± 4.7||20.4 ± 4.3||.14||22.1 ± 4.7|
Participants provided demographic information, health and injury history, and completed the Foot and Ankle Ability Measure (FAAM) ADL ( ) and Sport subscales ( ), Identification of Functional Ankle Instability (IdFAI) ( ), the Patient Reported Outcomes Measurement Information System (PROMIS) General Health Questionnaire ( ), the 11-item Tampa Scale of Kinesiophobia (TSK-11) ( ), and the Godin Leisure-time Exercise Questionnaire ( ). Height, mass, and leg length were measured. Foot posture was assessed in standing using the Foot Posture Index–6 item version (FPI), a categorical measure of foot type that is based on five observations and one palpatory assessment ( ).
Demographic, medical history, and FPI assessments were performed by a physical therapist and board-certified orthopaedic clinical specialist with 15-years of clinical experience. Physical examinations were performed by either an athletic trainer with three-years clinical experience or a physical therapist with two-years clinical experience who were blinded to participants’ medical history, functional status, and treatment allocation.
Morphologic foot assessment
Morphologic foot measurements were obtained using the Arch Height Index Measurement System (JAKTOOL Corporation, Cranberry, NJ). Total and truncated foot length, arch height, and foot width were measured in sitting and standing. Test-retest reliability for these measures were previously reported by the authors to be excellent ( ). Arch height flexibility ( ) and foot mobility magnitude ( ) were calculated using the component measurements across loading conditions.
A digital palpatory examination of the anterior talofibular (ATFL), calcaneofibular, and bifurcate ligaments was performed. Participants reported the presence or absence of pain.
Joint motion measures
Weight bearing dorsiflexion (WBDF) ( ), ankle plantarflexion, inversion, and eversion, and forefoot inversion and eversion joint motion measures ( ) were performed using previously described methods. WBDF was reported as the linear distance measured from the wall to the toes in centimeters. Joint motion measures of rearfoot plantarflexion, inversion, and eversion were performed using a 30.5-cm plastic goniometer (Merck Corporation, Kenilworth, NJ) and reported in degrees. Forefoot inversion and eversion was measured using a digital inclinometer (Fabrication Enterprises, White Plains, NY) and reported in degrees. Linear excursion of first metatarsal (MT) dorsiflexion and plantarflexion were measured utilizing a custom measuring device consisting of two metal rulers bent to 90° and reported in millimeters ( ). Test-retest reliability for these measures were previously reported by the authors to be good to excellent ( ; ).
Ankle dorsiflexion, plantarflexion, inversion, eversion, and hallux flexion and lesser toe flexion strength were assessed with a handheld dynamometer (Hoggan Health Industries, West Jordan, UT) ( ). For toe flexion strength measures, the ankle was positioned in 45° plantarflexion to reduce contribution of the extrinsic foot muscles and increase demand of the intrinsic foot muscles ( ). Strength measures were based on the highest value of three trials. An estimate of torque was derived from the product of force and segment length, normalized to body mass, and reported in Nmkg −1 . Test-retest reliability for these measures were previously reported by the authors to be excellent ( ).
Dynamic balance was assessed using the anterior, posteromedial, and posterolateral directions of the Star Excursion Balance Test (SEBT) ( ), a measure has been found to have excellent test-retest reliability ( ). Reach distance was normalized to leg length ( ). A composite measurement was calculated from the mean of the three directions.
Following baseline assessment, all participants were instructed in a HEP consisting of triceps surae stretching; four-way stretch of the rearfoot, midfoot, and forefoot; isotonic inversion, eversion and dorsiflexion exercises against resistance tubing; single-limb heel raising; and a single limb balance exercise ( Fig. 2 ). ( ) Participants were asked to perform all exercises three times daily, were provided a handout detailing the exercises, and verbalized understanding following instruction. The decision to utilize a HEP over supervised rehabilitation was to better elucidate the specific treatment effects of the midfoot joint mobilizations.
Participants were randomized a priori using a random number generator by the senior author and stratified by sex to receive either the midfoot joint mobilizations or sham intervention on the initial visit. Allocation was performed by an otherwise uninvolved laboratory assistant, concealed in a sealed opaque envelope, and opened by the treating clinician. Participants allocated to receive midfoot mobilizations were provided a dorsolateral cuboid glide with forefoot supination and 1st tarsometatarsal plantar glides ( ). Each mobilization technique was an oscillatory Maitland Grade IV applied for 30-seconds duration. If cavitation was not experienced during the first bout of oscillations, a second 30-second bout was provided. Participants allocated to the sham treatment were told that they were to receive a gentle soft-tissue technique similar to massage and were provided a “laying of hands” for 30-seconds using the same hand position and contacts used for the joint mobilizations. Participants rated the change of symptoms using a single assessment numeric evaluation (SANE, −100% = full exacerbation, 0 = no change, 100% = full resolution) immediately post-intervention and completed the Global Rating of Change (GROC, −7 = A very great deal worse, 0 = About the same, 7 = A very great deal better).
Participants returned to the laboratory following a one-week washout for reassessment. They completed the PROMIS, Godin, FAAM-ADL and Sport, SANE, and GROC. HEP compliance was assessed by having the participants demonstrate the instructed exercises. Participants were rated by the treating clinician whether or not they could demonstrate the exercises without hesitation and with appropriate technique. Participants rated their compliance using a SANE, with 0% reflective of complete non-compliance with all home exercises and 100% representing performance of all exercises three times daily. Any deficiencies in exercise technique were corrected and participants were provided encouragement to continue.
Repeat physical examinations were performed pre-and post-intervention. Following the pre-intervention physical examination, participants crossed over to receive the second intervention (i.e. individuals who initially received the sham intervention now received the midfoot joint mobilizations). Participants rated treatment response (SANE) immediately post-intervention and at the end of the visit and completed the GROC.
Participants returned to the laboratory one-week later for the final reassessment visit consisting of HEP compliance, patient-reported outcomes, and physical examination.
A priori sample size estimation of 14 participants were needed based on an anticipated 15-point change in the FAAM Sport, an <SPAN role=presentation tabIndex=0 id=MathJax-Element-1-Frame class=MathJax style="POSITION: relative" data-mathml='α’>𝛼α
=.05, and <SPAN role=presentation tabIndex=0 id=MathJax-Element-2-Frame class=MathJax style="POSITION: relative" data-mathml='β’>𝛽β
=.20 ( ). Descriptive statistics were calculated for demographic and self-reported measures for each subset of the sample allocated to receive either sham or midfoot mobilization during the first visit. Independent t-tests were used to assess differences between allocation groups.
Treatment effectiveness of the two interventions (midfoot joint mobilization, sham) on pain in the past week and pre-and post-intervention patient-reported outcomes measures was compared using dependent t -tests. Proportion estimates and 95% CI were calculated for dichotomous variables. The effects of treatment (midfoot joint mobilization, sham) and time (immediate change, pre-to-1-week change) on clinical measures were assessed using within-subjects repeated measures ANOVAs. Change scores that had a 95% CI that did not cross zero were considered to represent a significant change. Significant changes in outcomes observed immediate post-treatment that were non-significant pre-to-1-week were interpreted as having an immediate, but fleeting effect. Ordinal measures that had greater than five items (GROC) were treated as continuous data during analysis ( ; ). Participants were analyzed per allocation using intention to treat.
Post hoc Cohen’s d effect size (ES) point estimates and 95% CI 8 comparing treatments were calculated for all significant treatment or treatment by time interactions for immediate pre-to-post and 1-week change scores. ES 0–0.19 were interpreted as trivial, 0.2–0.49 small, 0.5–0.79 moderate, and >0.8 large ( ).
Data was analyzed using Statistical Package for Social Sciences (SPSS) Version 23.0 (IBM, Inc., Armonk, New York) and Microsoft Excel for Mac 2016 (Microsoft Corp., Redmond, WA). The level of significance was p ≤ .05 for all analyses.
There were no statistical differences in demographics, injury history, foot posture, or patient-reported outcome measures between allocation groups at initial baseline, with the exception of idFAI ( Table 1 ). The group allocated to receive the sham intervention first had significantly greater self-reported instability (IdFAI = 26.6 ± 3.5) compared to the joint mobilization first group (IdFAI = 20.7 ± 4.6, p = .009). Objective physical measures were similar between groups prior to treatment. ( Tables 3–5 ). In the assessment of carryover effects between baselines 1 and 2, rearfoot inversion motion was the only measure that had a significant treatment by order interaction.
|Baseline 1||Pre-to-Post Change||Baseline 2||Pre-to-Post Change|
|Mean ± SD||p|
|Pain VAS (cm)|
|At Present||Sham||1.4 ± 1.5 a||IMM:0.3 ± 1.8||0.9 ± 1.0 b||IMM:0.4 ± 0.6||.10 c |
|1wk: −0.2 ± 1.3||1wk: 0.4 ± 0.6|
|Treatment||2.0 ± 2.1 b||IMM: 0.6 ± 1.3||1.2 ± 1.7 a||IMM:0.2 ± 0.7|
|1wk: −1.1 ± 1.4||1wk: −0.8 ± 1.7|
|Worst in the Past Week||Sham||3.5 ± 1.9 a||−1.9 ± 1.7||1.8 ± 1.1 b||−0.7 ± 1.3||.10|
|Treatment||3.3 ± 1.8 b||−1.6 ± 1.6||1.6 ± 1.8 a||−0.9 ± 1.6|
|PROMIS General Health ( t )|
|Physical Composite||Sham||49.2 ± 5.8 a||2.9 ± 5.5||55.6 ± 3.9 b||2.4 ± 6.7||.87|
|Treatment||52.0 ± 6.8 b||2.4 ± 6.7||52.0 ± 6.0 a||1.8 ± 5.9|
|Mental Composite||Sham||55.9 ± 7.0 a||1.3 ± 5.2||61.7 ± 4.5 b||3.1 ± 2.1||.28|
|Treatment||62.1 ± 8.3 b||−2.8 ± 3.0||57.1 ± 11.0 a||1.9 ± 7.5|
|Godin Leisure Time Activity||Sham||47.1 ± 27.2 a||3.5 ± 13.5||84.3 ± 101.4 b||9.8 ± 19.5||.32|
|Treatment||42.8 ± 24.7 b||39.5 ± 92.7||50.6 ± 19.0 a||7.1 ± 13.4|
|ADL Score||Sham||74.0 ± 17.1 a||12.3 ± 16.1||92.6 ± 4.0 b||2.0 ± 3.9||.81|
|Treatment||78.7 ± 13.3 b||12.0 ± 10.5||86.3 ± 9.4 a||4.4 ± 4.8|
|Sport Score||Sham||46.9 ± 27.7 a||19.9 ± 22.0||70.5 ± 18.7 b||13.1 ± 16.0||.88|
|Treatment||46.6 ± 31.4 b||22.9 ± 21.3||66.8 ± 20.3 a||10.8 ± 11.3|
|Perceived Improvement SANE (%)||Sham||IMM:4.0 ± 12.7 a||IMM:10.0 ± 31.3 b||.04 c |
|1wk: 35.7 ± 34.1 a||1wk: 65.0 ± 29.8 b|
|Treatment||IMM:36.2 ± 42.5 b||IMM:21.4 ± 21.7 a|
|1wk: 56.9 ± 33.1 b||1wk: 56.9 ± 37.0 a|
|Global Rating of Change||Sham||IMM: 0.1 ± 1.4 a||IMM:.63 ± 2.0 b||.05 c |
|1wk: 2.0 ± 1.9 a||1wk: 4.1 ± 1.7 b|
|Treatment||IMM 2.4 ± 2.6 b||IMM:1.4 ± 2.3 a|
|1wk: 3.1 ± 2.0 b||1wk: 3.9 ± 2.4 a|