http://informahealthcare.com/dre ISSN 0963-8288 print/ISSN 1464-5165 online Disabil Rehabil, Early Online: 1–10 ! 2014 Informa UK Ltd. DOI: 10.3109/09638288.2014.935877

RESEARCH PAPER

Disabil Rehabil Downloaded from informahealthcare.com by Memorial University of Newfoundland on 08/01/14 For personal use only.

Effects of joint mobilization on chronic ankle instability: a randomized controlled trial David Cruz-Dı´az1, Rafael Lomas Vega1, Maria Catalina Osuna-Pe´rez1, Fidel Hita-Contreras1,2, and Antonio Martı´nez-Amat1,2 1

Department of Health Sciences, Faculty of Health Sciences, University of Jae´n, Jae´n, Spain and 2Department of Anatomı´a y Embriologı´a, Facultad de Medicina, Institute of Biopathology and Regenerative Medicine (IBIMER), University of Granada, Granada, Spain Abstract

Keywords

Purpose: To evaluate the effects of joint mobilization, in which movement is applied to the ankle’s dorsiflexion range of motion, on dynamic postural control and on the self-reported instability of patients with chronic ankle instability (CAI). Methods: A double-blind, placebocontrolled, randomized trial with repeated measures and a follow-up period. Ninety patients with a history of recurrent ankle sprain, self-reported instability, and a limited dorsiflexion range of motion, were randomly assigned to either the intervention group (Joint Mobilizations, 3 weeks, two sessions per week) the placebo group (Sham Mobilizations, same duration as joint mobilization) or the control group, with a 6 months follow-up. Dorsiflexion Range of Motion (DFROM), Star Excursion Balance Test (SEBT) and CAI Tool (CAIT) were outcome measures. A separate 3  4 mixed model analysis of variance was performed to examine the effect of treatment conditions and time, and intention-to-treat (ITT) analysis was applied to evaluate the effect of the independent variable. Results: The application of joint mobilization resulted in better scores of DFROM, CAIT, and SEBTs in the intervention group when compared with the placebo or the control groups (p50.001). The effect sizes of group-by-time interaction, measured with eta-squared, oscillated between 0.954 for DFROM and 0.288 for SEBT posteromedial distance. In within-group analysis, the manipulation group showed an improvement at 6 months follow-up in CAIT [mean ¼ 5.23, CI 95% (4.63–5.84)], DFROM [mean ¼ 6.77, CI 95% (6.45–7.08)], anterior SEBT [mean ¼ 7.35, CI 95% (6.59–8.12)], posteromedial SEBT [mean ¼ 3.32, CI 95% (0.95–5.69)], and posterolateral SEBT [mean ¼ 2.55, CI 95% (2.20–2.89)]. Conclusion: Joint mobilization techniques applied to subjects suffering from CAI were able to improve ankle DFROM, postural control, and self-reported instability. These results suggest that joint mobilization could be applied to patients with recurrent ankle sprain to help restore their functional stability.

Chronic ankle instability, dorsiflexion, dynamic postural control, manual therapy, self-reported instability History Received 28 October 2013 Revised 10 June 2014 Accepted 13 June 2014 Published online 3 July 2014

ä Implications for Rehabilitation  



Functional instability is a very common sequela in patients with CAI, resulting in reduced quality of living due to the limitations it imposes on daily life activities. The mobilization with movement technique presented by Mulligan, and based on the joint mobilization accompanied by active movement, appears as a valuable tool to be employed by physical therapists to restore ankle function after a recurrent ankle sprain history. ROM restriction, subjective feeling of instability and dynamic postural control are benefiting from the joint mobilization application.

Introduction Ankle sprain is the most common injury in the active population [1–3] and accounts for 22% of all sports injuries [4]. The incidence in the general population has been reported to be 600–700 cases per 100 000 people per year, generating an

Address for correspondence: Professor Antonio Martı´nez-Amat, Department of Health Sciences, Faculty of Health Sciences, University of Jae´n, E-23071 Jae´n, Spain. Fax: +34953012141. E-mail: amamat@ ujaen.es

economic cost of $4 billion in the US alone [3]. Ankle sprain has traditionally been considered an innocuous injury [2,5] although it is well documented that between 70% and 80% of all patients with a previous history of ankle sprain [2,4,6,7] developed residual symptoms such as ligament laxity, loss of proprioception, decreased range of motion (ROM), recurrent swelling, pain during activity, and feelings of ‘‘giving way’’ and of ankle instability [4,8,9]. Hertel et al. [10] defined Chronic Ankle Instability (CAI) as ‘‘repetitive bouts of lateral ankle instability resulting in numerous ankle sprains’’. This term has been widely used as a homogeneous entity [11], nevertheless CAI

Disabil Rehabil Downloaded from informahealthcare.com by Memorial University of Newfoundland on 08/01/14 For personal use only.

2

D. Cruz-Dı´az et al.

is a complex ankle disorder that maybe cause by functional instability (FI), mechanical instability (MI), or a combination of both conditions [11]. FI is defined as the subjective feeling of instability and is in relation with a proprioceptive and neuromuscular dysfunction while MI is more objective and involves the movement of the ankle joint beyond the physiologic ROM [10]. Patients with MI may control their ankle joint although they present more laxity than normal, being the feeling of ‘‘giving way’’ typical in CAI patients due to FI [10]. There are some predisposing factors to present CAI that could be classified as either intrinsic or extrinsic [12]. Increased ankle eversion to inversion strength ratio, plantarflexion strength, dorsiflexion to plantarflexion strength ratio, limb dominance, lower leg alignment, decreased ROM or postural control could be classified as intrinsic factors. Extrinsic factors include physical activity, type of ground, and type of shoes worn [13,14]. Patients looking for physical therapy to treat an ankle sprain are usually worried about pain and the eventual recovery of the ankle joint functionality. Conventional physical therapy techniques such as taping, electrotherapy, thermotherapy, etc., has shown good results in decreasing the edema associated with ankle sprain and its associated pain [15]. Nevertheless this pain improvement, does not provide a solution to the possible sequelae associated with ankle sprain, such as proprioception impairment, muscle weakness, alterations in the joint’s ROM, etc. For this reason it is important to make a integral approach to treating ankle sprains in order to avoid the risk of recurrence [4,16]. The relative risk factor of resprain is a previous episode of ankle sprain, and it is estimated at up to 70% [17]. Dorsiflexion range of motion (DFROM) is often affected after ankle sprain injury [18]. A lack of DFROM can predispose to re-injury and it has been included as a risk factor [6,19]. DFROM is associated with an alteration in normal talar arthrokinematics, being a reduced posterior talar glide or a positional alteration of the talus in relation with the ankle mortise two possible hypotheses [2]. Manual therapy approach, tend to lengthen the joint capsule and associated ligaments by stretching them through accessory motion in order to restore DFROM by increasing the extensibility of non-contractile tissues [2,10]. The role of the gastrocnemius and soleus muscles tightness has been deemed as a contributing factor for DFROM nevertheless following the results obtained by Johanson in 2008, DFROM seems to be more in relation with subtalar joint than muscle tightness [20]. Some manual therapy techniques based on talocrural joint mobilization have proven the effectiveness of this kind of stimuli in DFROM and arthrokinematic improvement [2,9,20,21]. Balance is usually altered in patients with CAI, especially during dynamic balance tasks [14,22]. This seems to be due to a deficit in proprioception and in the neuromuscular control of the injured ankle [11], and decreased balance is deemed to be a risk factor for recurrent ankle sprain [23]. Joint mobilizations are linked to positive effects in postural control, which is reported to be impaired in those suffering from CAI [2,6]. Mobilization with movement (MWM) was described by Mulligan [24] and is deemed to be effective in reducing pain and swelling, improving function, persistent instability feeling, postural control impairment, and ankle DFROM in lateral ankle sprain patients [21,25]. These symptoms are widely present, together or individually, in CAI cases in which conventional treatment has been inefficient. Joint mobilization techniques are effective in reducing pain, increased muscle tone and DFROM which is considered as a predisposing factor to re-injury and whose improvement is hypothesized to be derived by the posterior talar glide alteration [18,21,26]. The articular stretching due to joint mobilizations, increase the sensory output of mechanoreceptors in capsule and ligaments due to the activation of gamma

Disabil Rehabil, Early Online: 1–10

motor neurons by tissue traction which is related with postural control improvement [27]. As described by several authors [24,27], MWM can be employed in acute and sub-acute ankle sprain in weight-bearing and non-weight-bearing conditions with the particularity that the patient moves actively. Some of its advantages have been described in many publications, such as the immediate and rapid improvement in pain and movement, and in addition the use of joint mobilization techniques is increasingly extended [4] due to evidence-based clinical practice. However, the progressively cumulative effect over a number of sessions has not been documented, as stated by Reid et al. There is a lack of controlled clinical trials focusing on this cumulative effect [21], defined as the amount of improvement due to continuous treatment sessions, especially regarding ankle instability. Further studies are needed to determine the evolution of patients with CAI, during and after treatment, in the variables under study here. The purpose of this randomized clinical trial was to determine the short- and long-term effects (in a 6-month follow-up) of 3 weeks of MWM on DFROM, dynamic postural control, and self-reported instability of patients with CAI.

Methods Participants A randomized double-blind placebo-controlled trial was conducted. A total of 102 patients were screened for inclusion in the study. To be accepted in this study participants had to meet the following inclusion criteria: (1) a previous history of ankle sprain with at least two sprains on the same side in the last 2 years; an asymmetry larger than 2 cm on the weight-bearing lunge test (WBLT) for ankle dorsiflexion; (3) no history of lateral ankle sprain on the contralateral side; (4) self-reported instability and feeling of ‘‘giving way’’; and (5) not receiving any other physical therapy treatment during the study. Exclusion criteria included: (1) acute ankle sprain within the previous 6 months; (2) a history of bilateral ankle injury; (3) bony injury associated with ankle sprain such as avulsion fracture or ankle osteochondral lesion; and (4) previous injury or surgery to the back, hip, or knee. Participants were excluded from the study if they missed more than two therapeutic sessions. The study was approved by the Human Ethics Committee of the University of Jae´n. Informed consent was obtained from all participants and the rights of the participants were protected. Ninety subjects were included and randomized to each group. The flowchart of patient recruitment and retention can be found in Figure 1. Finally, eighty-one subjects (47 male and 34 female, mean age of 27.7 years and SD ¼ 6.80) completed the study and were analyzed. Outcome measures (dependent variables) Weight-bearing ankle DFROM WBLT assesses the maximal advancement of the tibia over the talus in a weight-bearing position, and it was performed to determine ankle DFROM [8,12,21]. This test is widely used to assess DFROM in the ankle joint due to its excellent psychometric properties. This test is more reliable (ICC ¼ 0.93–0.96) than measures obtained in a non-weight-bearing position and has been applied on patients with CAI [28], where it has proven to correlate with dynamic postural control measures [18]. Subjects performed three practice and three analysis trials of the test on the involved limb. The average of the three analysis trials was calculated and used for statistical analysis. Patients were placed in front of a wall and told to place their big toe in line with the heel and on top of a white tape mark on the floor. Keeping the heel firmly on the ground, patients were instructed to bend the supporting knee so that it touched the wall. Thus, the maximum distance the subject

DOI: 10.3109/09638288.2014.935877

Joint mobilization on chronic ankle instability

3

Disabil Rehabil Downloaded from informahealthcare.com by Memorial University of Newfoundland on 08/01/14 For personal use only.

Figure 1. Flowchart of patient recruitment and retention. CAIT, Cumberland Ankle Instability Tool; ROM, range of motion.

can place the foot away from the wall while keeping both the heel flat on the floor and the knee touching the wall is measured in millimeters between the part of the foot that is closest to the wall and the wall itself [2,5,8]. Some authors recommend the use of weight-bearing measurement of dorsiflexion due to its ability to detect treatment effects better than non-weight-bearing measurement [27].

choose this questionnaire over others such as the Foot and Ankle Disability Index (FADI) or the Foot and Ankle Ability Measure (FAAM). Furthermore, until recently CAIT has been the only cross-cultural self-report questionnaire adapted into Spanish, unlike the tools that have been mentioned above and it presented good psychometrics properties (Cronbach’s  ¼ 0.766 and ICC 0.979 (95 % CI ¼ 0.958–0.99) [30–32].

Self-reported ankle instability

Dynamic postural control

Self-reported ankle instability, or the feeling of ‘‘giving way’’, is commonly reported in those with CAI [11]. The Cumberland Ankle Instability Tool (CAIT) is a nine-item questionnaire designed to evaluate several aspects of CAI. The total score of the nine items ranges from 0 (severe instability) to 30 (normal stability) where scores 27 indicate FI. The CAIT is a discriminative questionnaire that can identify patients with CAI and measure the severity of functional ankle instability [22,29]. A patient with a score 427 could be considered as non-affected ankle while punctuation close to 0 are related with a very severe affectation, being this instability classification better with higher scores close to 27 although these patients are diagnosed as CAI. This ability to grade and discriminate patients with CAI along with a reported capacity to detect changes over time made us

The Star Excursion Balance Test (SEBT) has been reported to be a highly reliable and valid tool to assess dynamic balance in those with CAI [13]. It consists of a single-limb stance with maximum reach of the opposite leg where shorter maximal reach distances are correlated with a poor postural control. Some authors advocate that the measure of the normalized anterior (ANT), the posteromedial (PM), and the posterolateral (PL) reach distances is correlated with CAI [5,18,]. The SEBT appears to have the sensitivity to detect reach deficits both between subjects and within sides of the same subject with unilateral ankle instability [13,22,33–35]. For the SEBT, patients stood on a single leg with the involved limb placed at the center of a grid, and maintained a single-limb stance with both hands placed on the hips while trying to reach the furthest point possible in anterior

4

D. Cruz-Dı´az et al.

Disabil Rehabil Downloaded from informahealthcare.com by Memorial University of Newfoundland on 08/01/14 For personal use only.

(ANT), posteromedial (PM), and posterolateral (PL) reach distances with the most distal part of the reach foot, while keeping slight toe contact with the tape measure [5,6,9,18,35]. Patients performed three trials in each direction for analysis, and the distance from the center of the grid to the touch point was manually measured with a tape measure. The reaches in each direction were normalized with the leg length. Lower limb length was determined, with the subject lying supine, by measuring the distance from the anterior superior iliac spine to the distal end of the medial malleolus. The SEBT composite score was calculated by dividing the sum of the three reach distances in the anterior (A), posteromedial (PM), and posterolateral (PL) directions by three times the limb length (LL) of the individual, then multiplied by 100 {[(A + PM + PL)/(LL  3)]  100}. Results were obtained in the form of a percentage [6,35,36]. Independent variables The independent variables used in this study were treatment (consisting of a weight-bearing MWM; MWM_WB) for the manipulation group, a sham treatment for the placebo group, and no treatment for the control group. For the MWM_WB, patients were instructed to stand in a high kneeling position with the affected ankle in a weight-bearing neutral position and one upperextremity support for stability. A rigid belt was passed around the inferior margin of the medial malleolus and the waist of the therapist while manually fixing the talus and calcaneus with both hands. Thus, the therapist drew the tibia forward on the talus, resulting in a relative posterior talar glide of the talus in the ankle mortise. Holding this sustained postero-anterior glide to the tibia through the belt, patients were told to perform a slow dorsiflexion movement until the first onset of pain or until the end of their range. According to the ‘‘no pain rule’’ described by Mulligan [24], if the procedure was painful for the patient, the therapist readjusted the contact to avoid any pain. Two sets of 10 repetitions, separated by a 2-min rest, were performed as described by Mulligan [24]. In the sham intervention the objective was to fix the talocrural joint. Thus, the therapist placed a semi-rigid orthesis that limited ankle dorsiflexion. With the patient in a supine position, the therapist passively flexed and extended the knee of the patient for two sets of the 10 repetitions following the same protocol as in the manipulation group. A no-treatment control condition was included in this study for comparison with the results obtained in the intervention and in the placebo groups. The intervention and the placebo groups received two treatment sessions per week during a 3-week period. Experimental procedure Participants who met the inclusion criteria and accepted to be enrolled in this study were required to report the laboratory for testing. The subjects’ ankle ROM, dynamic postural control, and self-reported instability feeling were then assessed in the same order for all subjects by a blinded examiner who did not know the group allocation of patients. Both patients and investigators were requested to refrain from discussing treatment. These measurements were repeated just after the first joint mobilization session, after the completion of the 3 weeks of treatment, and after 6 months from the end of the intervention. Self-reported instability was not measured after the first session. No warm-up or stretching activities preceded the measurement. Participants were instructed to maintain normal physical and activities of daily life (ADLs) and to avoid any alternative treatment during the length of the study. Randomization was accomplished through the use of a computer-generated table of random numbers. Patients were then assigned to the intervention, the placebo, or the control

Disabil Rehabil, Early Online: 1–10

groups. Allocations were sealed in opaque and consecutivelynumbered envelopes kept at a locked location. These were opened in sequence by an independent administrator not involved in eligibility assessment, outcome assessment, or treatment. Allocation was revealed to the treating physical therapist via e-mail before participants arrived for experimental intervention. Sample size determination The sample size was performed using Ene 3.0 (Autonomous University of Barcelona, Spain). The required sample was determined taking as a reference the data reported by Hoch MC et al. [2]. For a minimal detectable change of 4.28 in the posterolateral SEBT, with a power of 0.80 and a significance level of 95%, 21 subjects per group are required. For a medium-term follow-up with 20% expected losses [13], five additional subjects per group were required. The total number of subjects required was 78. A total of 90 subjects took part in the study. Data analysis Data were analyzed using SPSS 17.0 (SPSS, Inc, Chicago, IL) and MedCalc 12.5 (MedCalc, Mariakerke, Belgium). The description of continuous variables was performed through mean and standard deviations, and the description of categorical variables through frequencies and percentages. In order to test baseline comparability between groups, baseline and socio-demographic data were analyzed with a one-way analysis of variance for the continuous variables and with a chi-squared test for the categorical variables. A 3  3 mixed-model analysis of variance (ANOVA) was used to examine the effect of treatment conditions (control, sham, or intervention) as the between-subjects variable, and time (pretreatment, 3 weeks, and 6 months) on CAIT scores as the within-subject variable. A separate 3  4 mixed model ANOVA was carried out to examine the effect of treatment conditions and time (pretreatment, post-first-session, post3-weeks of treatment, and 6 months follow-up) on the dependent variables (ROM, SEBTs). The hypothesis of interest was the group by-time interaction at an alpha level of 0.05. For the determination of effect size of the group-by-time interaction, eta-squared was used. Additionally, if a significant interaction was identified, pairwise Bonferroni comparisons were performed to explore the differences between groups and between time points. Two separate analyses were performed to evaluate the effect of the independent variable: (1) ITT analysis with all participants that were allocated to each group condition, and (2) a per protocol analysis of the data of all participants who completed the followup. For the ITT analysis, expectation maximization (EM) was used for the estimation of missing values.

Results A total of 81 patients (47 males, 34 females, mean age of 27.7 years and SD ¼ 6.80) completed the study and were analyzed (Figure 1). They were included in the manipulation group (n ¼ 30; 56.7% males), the placebo group (n ¼ 31; 54.8% males), or the control group (n ¼ 29; 58.6% males). At baseline, the groups were similar for all dependent and socio-demographic variables in perprotocol analysis. On ITT analysis (n ¼ 90) no differences were observed between groups at baseline (Table 1). In the analysis of missing values, Little’s MCAR test showed a chi-squared ¼ 62.780 (p ¼ 0.050), so it cannot be ruled out that the values were missing completely at random (MCAR). The overall group-by-time interaction for each mixed-model ANOVA is shown in Table 2. The interaction was statistically significant at an alpha level50.001 for all dependent variables in ITT analysis.

Joint mobilization on chronic ankle instability

DOI: 10.3109/09638288.2014.935877

5

Disabil Rehabil Downloaded from informahealthcare.com by Memorial University of Newfoundland on 08/01/14 For personal use only.

Table 1. Baseline and socio-demographic data on intention-to-treat (ITT) analysis.

Age (years) Height (m) Weight (kg) BMI (kg/m2) CAIT pre ROM pre SEBT ANT Pre SEBT PM Pre SEBT PL Pre Gender Male Female Ankle Left Right

Control (n ¼ 29)

Sham (n ¼ 31)

Intervention (n ¼ 30)

p Value

26.48 ± 4.03 1.70 ± 0.08 68.57 ± 10.51 23.75 ± 2.63 22.31 ± 2.27 53.33 ± 2.00 78.70 ± 5.77 84.20 ± 7.02 87.04 ± 3.21

29.55 ± 9.44 1.72 ± 0.07 69.02 ± 9.59 23.19 ± 2.15 21.81 ± 1.80 53.72 ± 1.68 77.21 ± 5.84 85.59 ± 3.04 86.62 ± 3.14

26.83 ± 4.62 1.71 ± 0.09 68.97 ± 10.10 23.57 ± 2.88 21.23 ± 1.25 53.99 ± 1.58 77.88 ± 5.02 85.18 ± 6.42 86.25 ± 2.88

0.143 0.636 0.982 0.693 0.080 0.365 0.583 0.633 0.617

17/58.6% 12/41.4%

17/54.8% 14/45.2%

17/56.7% 13/43.3%

0.957

10/34.5% 19/65.5%

11/35.5% 20/64.5%

9/30.0% 21/70.0%

0.891

BMI, body mass index; CAIT, Cumberland Ankle Instability Tool; ROM, range of Motion; SEBT, star excursion balance test. Data are provided as means and SD for continuous variables, and as frequencies and percentages for categorical variables. Analysis of variance was used for continuous variables, and the chi-squared test was used for categorical variables. Table 2. Descriptive data, statistical significance and effect size of the group-by-time interaction on ITT analysis.

Pre-treatment CAIT Control Sham Intervention ROM Control Sham Intervention SEBT Ant Control Sham Intervention SEBT PM Control Sham Intervention SEBT PL Control Sham Intervention

After first session

22.31 ± 2.27 21.81 ± 1.80 21.23 ± 1.25

After 3 weeks of treatment

6 month post-treatment follow-up

22.42 ± 2.31 22.17 ± 1.85 27.87 ± 1.17

22.42 ± 2.55 22.02 ± 2.08 26.47 ± 1.48

53.33 ± 2.00 53.72 ± 1.68 53.99 ± 1.58

53.41 ± 1.92 53.83 ± 1.71 60.54 ± 1.67

53.54 ± 1.87 54.01 ± 1.84 60.84 ± 1.72

53.49 ± 1.97 53.80 ± 1.94 60.75 ± 1.69

78.70 ± 5.77 77.21 ± 5.84 77.88 ± 5.02

78.57 ± 6.07 77.09 ± 6.01 84.72 ± 5.07

78.69 ± 6.10 78.14 ± 6.38 86.24 ± 5.20

78.81 ± 6.07 77.44 ± 6.20 85.23 ± 5.10

84.20 ± 7.02 85.59 ± 3.04 85.18 ± 6.42

82.36 ± 7.17 85.52 ± 2.86 88.51 ± 5.43

79.22 ± 5.48 85.32 ± 2.80 89.44 ± 5.46

79.37 ± 5.58 85.48 ± 2.76 88.50 ± 5.36

87.04 ± 3.21 86.62 ± 3.14 86.25 ± 2.88

87.30 ± 3.23 87.11 ± 3.25 89.28 ± 3.04

87.49 ± 3.12 87.22 ± 3.29 89.77 ± 2.96

87.53 ± 3.21 86.58 ± 3.30 88.80 ± 2.94

Group  Time

Eta squared

p50.001

0.753

p50.001

0.954

p50.001

0.821

p50.001

0.288

p50.001

0.731

CAIT, Cumberland Ankle Instability Tool; ROM, range of motion; SEBT, star excursion balance test. Data are provided as means and SD.

In per-protocol analysis, all variables showed statistically significant group-by-time interaction (p50.001) and effect sizes measured with eta-squared were 0.753 (CAIT), 0.954 (ROM), 0.821 (SEBT ANT, 0.288 (SEBT PM) and 0.731 (SEBT PL). Effect sizes were similar in ITT and per-protocol analyses. CAIT Regarding the between-group analysis (Figure 2A), pairwise comparisons of Bonferroni showed significant differences between the manipulation group and the other two groups at the end of the 3-week treatment period and after the 6 months followup (p50.001). At 3 weeks, the manipulation group showed a better CAIT score than the sham group [Mean Difference ¼ 5.70, 95% CI (4.55–6.84)] and the control group [Mean Difference ¼ 5.45, 95% CI (4.29–6.61). At the 6 months followup, the manipulation group showed a better score than the sham group [Mean Difference ¼ 4.45, 95% CI (3.15–5.75)] and the control group [Mean Difference ¼ 4.05, 95% CI (2.73–5.37)].

Regarding within-group change scores (Table 3), a significant increase of CAIT scores was observed at 3 weeks and at 6 months (p50.001) in the manipulation group. ROM Regarding the between-group analysis (Figure 2B), pairwise comparisons of Bonferroni showed significant differences at the three assessments of the follow-up between the manipulation group and the other groups (p50.001). After the first manipulation session, the manipulation group showed a better ROM score than the sham group [Mean Difference ¼ 6.71, CI 95% (5.60–7.82)] and the control group [Mean Difference ¼ 7.13, CI 95% (6.00–8.25)]. At 3 weeks, the mean differences between the intervention group and the sham and control groups were, respectively, 6.84 [95% CI (5.79–7.97) and 7.30 [95% CI (6.15–8.45)]. At 6 months, the mean differences between the treatment group and the sham and control groups were 6.96 [95% CI (5.79–8.13)] and 7.27 [95% CI (6.08–8.45)]. As regards the within-group change scores, a significant increase (p50.001) was

6

D. Cruz-Dı´az et al.

Disabil Rehabil, Early Online: 1–10

Disabil Rehabil Downloaded from informahealthcare.com by Memorial University of Newfoundland on 08/01/14 For personal use only.

Figure 2. CAIT, Cumberland Ankle Instability Tool. ROM, Range of Motion. ***p50.001.

observed in the manipulation group in all time points of the follow-up (Table 3). Anterior SEBT Regarding the between-group analysis (Figure 3A), pairwise comparisons of Bonferroni showed significant differences at the three assessments of the follow-up between the manipulation group and the other groups (p50.001). After the first manipulation session, the manipulation group showed a better anterior SEBT score than the sham group [Mean Difference ¼ 7.62, 95% CI (4.04–11.21)], and the control group [Mean Difference ¼ 6.15, 95% CI (2.50–9.79)]. At 3 weeks, the mean differences between the manipulation group and the sham group were 8.10 [95% CI (4.40–11.80)] and 7.56 [95% CI (3.79–11.32)]. At 6 months, the mean differences between the treatment group and the sham and control groups were 7.79 [95% CI (4.16–11.43)] and 6.42 [95% CI (2.73–10.12)] respectively. Regarding the within-group change scores, the manipulation group exhibited a significant increase in all follow-up points (p50.001). The details of the rest of timepoint differences are found in Table 3. Posteromedial SEBT Regarding between-group analysis (Figure 3B), pairwise comparisons of Bonferroni showed, right after the first session, that

significant differences appeared between the manipulation group and the control group [Mean Difference ¼ 6.16, 95% CI (2.72– 9.59), p50.001]. At 3 weeks of treatment there were significant differences between the manipulation and the control group [Mean Difference ¼ 10.22, 95% CI (7.22-13.22), p50.001] and between the manipulation and the sham group [Mean Difference ¼ 4.13, 95% CI (1.17–7.08), p ¼ 0.003]. After 6 months since the end of treatment there were significant differences between the manipulation group and the control group [Mean Difference ¼ 9.12, 95% CI (6.13–12.12), p50.001], between the experimental and the sham group [Mean Difference ¼ 3.02, 95% CI (0.07–5.97), p ¼ 0.043], and between the sham and the control group [Mean Difference ¼ 6.11, 95% CI (3.13–9.08), p50.001]. The manipulation group showed a significant increase at the end of treatment and at 3 weeks (p50.001) as well as at the 6 months follow-up (p ¼ 0.002). However, the control group showed a statistically significant worsening at 3 weeks and at the 6 months follow-up (p50.001). Posterolateral SEBT In the between-group analysis (Figure 3C), pairwise comparisons of Bonferroni showed that after first manipulation session there were significant differences between the manipulation group and the sham group [Mean Difference ¼ 2.17, 95% CI (0.18–4.15), p ¼ 0.028]. At 3 weeks after treatment there were significant

Joint mobilization on chronic ankle instability

DOI: 10.3109/09638288.2014.935877

7

Table 3. Within-group change scores at the follow-up on ITT analysis. Within- Group change scores after first session

Disabil Rehabil Downloaded from informahealthcare.com by Memorial University of Newfoundland on 08/01/14 For personal use only.

Mean Diff CAIT Control Sham Intervention ROM Control Sham Intervention SEBT Ant Control Sham Intervention SEBT PM Control Sham Intervention SEBT PL Control Sham Intervention

CI 95%

Within-group change scores at the end of treatment (3 weeks) p Value

Mean Diff

CI 95%

p Value

0.11 0.36 6.63

(0.44, 0.66) (0.17,0.89) (6.09,7.17)

1.000 0.299 50.001

Within-group change scores at the follow-up (6 months) Mean Diff

CI95%

p Value

0.11 0.21 5.23

(0.51, 0.72) (0.38, 0.80) (4.63,5.84)

1.000 1.000 50.001

0.08 0.11 6.55

(0.11, 0.26) (0.07, 0.29) (6.37, 6.73)

1.000 0.569 50.001

0.21 0.29 6.86

(0.04, 0.46) (0.05, 0.54) (6.61, 7.11)

0.164 0.011 50.001

0.16 0.08 6.77

(0.17, 0.48) (0.23, 0.39) (6.45, 7.08)

1.000 1.000 50.001

0.13 0.11 6.84

(0.67, 0.41) (0.63, 0.41) (6.31, 7.37)

1.000 1.000 50.001

0.01 0.94 8.36

(0.83, 0.81) (0.15,1.73) (7.56,9.17)

1.000 0.012 50.001

0.11 0.24 7.35

(0.67, 0.89) (0.52,0.99) (6.59, 8.12)

1.000 1.000 50.001

1.85 0.07 3.33

(3.64, 0.05) (1.80, 1.67) (1.57, 5.10)

0.040 1.000 50.001

4.98 0.27 4.27

(7.37, 2.59) (2.59, 2.04) (1.91, 6.62)

50.001 1.000 50.001

4.83 0.11 3.32

(7.24, 2.42) (2.44, 2.22) (0.95, 5.69)

50.001 1.000 0.002

(0.06, 0.58) (0.18, 0.80) (2.72, 3.34)

0.185 50.001 50.001

0.45 0.60 3.52

(0.10, 0.79) (0.27, 0.94) (3.18, 3.86)

0.004 50.001 50.001

0.49 0.04 2.55

(0.14, 0.85) (0.38, 0.30) (2.20, 2.89)

0.002 1.000 50.001

0.26 0.49 3.03

CAIT, Cumberland Ankle Instability Tool; ROM, range of motion; SEBT, star excursion balance test.

differences between the manipulation group and the control group [Mean Difference ¼ 2.28, 95% CI (0.30–4.27), p ¼ 0.019] and between the manipulation and the sham group [Mean Difference ¼ 2.55 95% CI (0.59–4.50), p ¼ 0.006]. At 6 months of the end of treatment there were significant differences between the manipulation group and the sham group [Mean Difference ¼ 2.21, 95% CI (0.24–4.19), p ¼ 0.022] As regards the within-group change scores, at 6 months of follow-up the control group (p ¼ 0.002) and the manipulation group (p50.001) showed a significant increase. Details for each time point and each group are displayed in Table 3.

Discussion The present study deems joint MWM a useful therapeutic tool that provides good results in DFROM, dynamic postural control, and self-reported feelings of instability. It was observed that there is a deficit in DFROM in patients with CAI, which could be associated with functional impairments such as sensorimotor alterations, subjective feeling of ‘‘giving way’’, muscle activation, etc [4,27]. The influence of DFROM is deemed to be due to the alteration of normal arthrokinematics of the ankle as a consequence of the joint disrupting the normal transmission of afferent information available to the sensorimotor system [18]. Therefore, DFROM restoration should be a priority in the design of CAI treatment. The MWM_WB described by Mulligan was applied to the manipulation group and showed an immediate effect in DFROM when compared to the sham or the control groups (Tables 2 and 3). These findings are similar to those reported in other studies where joint mobilization is used [2,21,27]. Several authors have highlighted the possible cumulative treatment effect caused by the continuous application of joint mobilization over time, but there is no available literature concerning this issue [4,21,27,33,37]. On the other hand, the effectiveness of MWM has been reported in acute and subacute ankle sprain [37,38] and some articles have noted the application benefits of this technique in patients suffering from CAI [4,33,38,39]. Taking into account the outcome measures of the studies mentioned above, findings regarding MWM are similar,

and it seems to constitute an effective treatment tool for ankle sprain and ankle instability. The role of subtalar joint in arthrokinematics is considered as a factor in the development of functional and sensorimotor impairments [18], so it is possible that joint mobilization could restore the normal joint’s ROM and relieve the negative effects in afferent signals as well as the proprioceptive alterations. For this reason, we believe that MWM can be a good form of joint mobilization since it includes the particularity of the patient actively moving while the therapist performs the joint mobilization, unlike other forms of mobilization which require the therapist to passively manipulate the patient [24]. In the current study, we intended to analyze not only the immediate post-session improvement in DFROM but also the cumulative effect after 3 weeks of MWM. Results suggest that major changes occurred after the first session and that a moderate, although clinically relevant, effect happened during the 3 weeks of MWM when compared with the sham group or the control group (Tables 2 and 3). These results seem reasonable if we take into account that patients with greater DFROM restrictions would experience greater gains [21]. The effectiveness of a therapeutic technique should be considered not only under the perspective of its immediate post-application effect. The ability to maintain the gains obtained over a certain length of time should play a useful role when recommending the use of a therapeutic tool in evidence-based treatment guides. Nevertheless, the lack of moderately long follow-up studies makes it difficult to evaluate this effect. A 6 months follow-up process was carried out for the population under study. Subjects in the manipulation group experienced a slight decrease in DFROM after the 3-week treatment period, but results show that, as an average, there was a gain in DFROM 6 months after the intervention. Data obtained in the study indicate that DFROM values after 6 months were higher than those measured after the first treatment session. This leads us to think that the cumulative effect of the MWM treatment might be responsible for this increase (Tables 2 and 3). Although major changes happened just after the first session, it was observed that the continuous application of MWM achieved better results in the variables under study. The gains were minor, but constant, in comparison to the first manipulation. It is unclear if the proximity

8

D. Cruz-Dı´az et al.

Disabil Rehabil, Early Online: 1–10

Disabil Rehabil Downloaded from informahealthcare.com by Memorial University of Newfoundland on 08/01/14 For personal use only.

Figure 3. SEBT, Star Excursion Balance Test. *p50.05, **p50.01, ***p50.001, between the manipulation group and the other two groups. Framed asterisks show statistical differences between the two groups connected by vertical lines.

of the treatment session application could have a residual positive effect in the next session, but the improvements achieved endured. Decreased postural control is widely considered a consequence of previous ankle sprains, and correlates with CAI [10,13,15,35]. Joint mobilization is hypothesized to effectively promote the activity of mechanoreceptors due to the stretching performed in the capsule and ligaments of the ankle, which increase their sensory output as gamma motor neurons activate with this tissue traction [32,38]. Dynamic postural control improvement in

patients with CAI has been examined through SEBT assessment on this study. This is a dynamic balance measurement tool that has been reported to be more appropriate than center-of-pressure and excursion velocity, since these have not consistently detected postural control deficits associated with CAI. For these reasons, its use is recommended as it is more reliable and valid for the detection of significant improvements associated with rehabilitation [27,29,40]. Balance training based on neuromuscular and proprioception tasks has been used in some randomized

Joint mobilization on chronic ankle instability

Disabil Rehabil Downloaded from informahealthcare.com by Memorial University of Newfoundland on 08/01/14 For personal use only.

DOI: 10.3109/09638288.2014.935877

controlled trials, where it has proven to improve balance and the severity of chronic instability [5,15,41]. Nevertheless, the available literature about joint mobilization and its influence in postural control in patients with CAI is limited when treating to explain the mechanism by which this technique is effective [18,21,33,42] until the study published by Riemann et al. [25] who focus on the changes in gamma motor neurons activation due to joint mobilizations. Outcome measures in this study confirmed a significant amelioration of all three SEBT reach distances in the manipulation group with respect to the control and the sham groups, and a slight but significant improvement in the sham group with respect to controls. These results match those of previous research based on the application of various joint mobilization techniques [12,39] but differ from other studies in which it is argued that a single joint mobilization was unable to change SEBT reach distances in the presence of modest increases in DFROM [2,12,18]. A possible restoration of joint arthrokinematics after joint mobilization could explain the DFROM gains and the subsequent improvement in others variables affected in those with CAI, such as dynamic postural control, proprioception, and the self-reported feeling of ‘‘giving way’’. According to a hypothesis that suggests that the anterior reach direction is more correlated with DFROM than the rest [43], our findings support this statement after considering the results obtained in both variables under analysis [15]. An improvement in all reach directions of SEBT was observed after the 3-week treatment period, which could be linked to the increase in DFROM. DFROM scores after the 6 months follow-up are slightly lower than after only 3 weeks of treatment. However, the results from both tests (DFROM and SEBT) are within a similar range, which in any case is over baseline scores. The slight improvement in the sham group could be attributed to a possible learning effect. The intervention showed an improvement in the self-reported feeling of instability measured by CAIT. These results support the hypothesis of the influence of joint mobilization in the proprioception achievement and the relationship between subtalar glide ROM restoration and sensorimotor recovery [18]. The intervention group experienced a significant improvement in CAIT scores after the completion of the 3-week treatment. Self-reported feelings of a more stable ankle increased in subjects who reported an increased DFROM and better SEBT scores. Some patients in the intervention group were identified as CAI at baseline, and after joint mobilization they obtained scores higher than 27, which could be considered as stable [10]. Nevertheless, the control and sham groups maintained similar scores during the 3 weeks of treatment and at the follow-up measurement (Table 2). A slight decrease in CAIT results was observed after the 6 months follow-up period, but better results were maintained than had been obtained at baseline. Thus, it could be hypothesized that MWM could enhance the subjective feeling of stability. This study has shown the long term effects of WB_MWM joint mobilization on patients with CAI. Post-immediate effects of joint manipulation have been widely described, but few authors have dealt specifically with the enduring improvements. The results obtained in the different variables during the follow-up point at the benefits of continued joint mobilization treatment.

Conclusion The application of the WB_MWM joint mobilization seems to be effective in the treatment of DFROM, dynamic postural control, and in self-reported instability. The treatment effects of this joint mobilization technique endure over time as observed in the follow-up. The influence of subtalar joint arthrokinematics and the restoration of its ROM on proprioception and the sensorimotor system should be considered in further studies.

9

Acknowledgements The authors wish to express their gratitude to the Medicarim Centro for their logistic support, to Ms M.a Del Pilar Cruz-Dı´az and Mr. Agustı´n Pe´rez-Barroso for their help in technical as well as practical issues.

Declaration of interest We certify that no party having a direct interest in the results of the research supporting this article has or will confer a benefit on us or on any organization with which we are associated and, if applicable, we certify that all financial and material support for this research (e.g. NIH or NHS grants) and work are clearly identified in the title page of the article.

References 1. Beynnon BD, Murphy DF, Alosa DM. predictive factors for lateral ankle sprains: a literature review. J Athl Train 2002;37:376–80. 2. Hoch MC, Anreatta RD, Mullineaux DR, et al. Two-week joint mobilization intervention improves self-reported function, range of motion, and dynamic balance in those with chronic ankle instability. J Orthop Res 2012;30:1798–804. 3. Waterman BR, Owens BD, Davey S, et al. The epidemiology of ankle sprains in the United States. J Bone Joint Surg Am 2010;92: 2279–84. 4. Fong DT, Hong Y, Chan LK, et al. A systematic review on ankle injury and ankle sprain in sports. Sports Med 2007;37:73–94. 5. Hale SA, Hertel J, Olmsted-Kramer LC, et al. The effect of a 4-week comprehensive rehabilitation program on postural control and lower extremity function in individuals with chronic ankle instability. J Orthop Sports Phys Ther 2007;37:303–11. 6. Gribble PA, Hertel J, Denegar CR, Buckley WE. The effect of fatigue and chronic ankle instability on dynamic postural control. J Athl Train 2004;39:321–9. 7. Holmes A, Delahunt E. Treatment of common deficits associated with chronic ankle instability. Sports Med 2009;39:207–24. 8. Bennell KL, Talbot RC, Wajswelner H, et al. Intra-rater and interrater reliability of a weight-bearing lunge measure of ankle dorsiflexion. Aust J Physiother 1998;44:175–80. 9. Green T, Refshauge K, Crosbie J, Adams R. A randomized controlled trial of a passive accessory joint mobilization on acute ankle inversion sprains. Phys Ther 2001;81:984–94. 10. Hertel J. Functional anatomy, pathomechanics, and pathophysiology of lateral ankle instability. J Athl Train 2002;37:364–75. 11. Hiller CE, Kilbreath SL, Refshauge KM. Chronic ankle instability: evolution of the model. J Athl Train 2011;46:133–41. 12. Hubbard TJ, Hertel J. Mechanical contributions to chronic lateral ankle instability. Sports Med 2006;36:263–77. 13. Olmsted LC, Carcia CR, Hertel J, Shultz SJ. Efficacy of the Star Excursion Balance Tests in detecting reach deficits in subjects with chronic ankle instability. J Athl Train 2002;37:501–6. 14. McKeon PO, Hertel J. Systematic review of postural control and lateral ankle instability: part 1. Can deficits be detected with instrumented testing? J Athl Train 2008;43:293–304. 15. Terada M1, Pietrosimone BG, Gribble PA. Therapeutic interventions for increasing ankle dorsiflexion after ankle sprain: a systematic review. J Athl Train 2013;48:696–709. 16. Pope R, Herbert R, Kirwan J. Effects of ankle dorsiflexion range and pre-exercise calf muscle stretching on injury risk in Army recruits. Aust J Physiother 1998;44:165–72. 17. Hubbard TJ, Kramer LC, Denegar CR, Hertel J. Contributing factors to chronic ankle instability. Foot Ankle Int 2007;28: 343–54. 18. Hoch MC, Staton GS, Medina McKeon JM, et al. Dorsiflexion and dynamic postural control deficits are present in those with chronic ankle instability. J Sci Med Sport 2012;15:574–9. 19. McHugh MP, Tyler TF, Tetro DT, et al. Risk factors for noncontact ankle sprains in high school athletes: the role of hip strength and balance ability. Am J Sports Med 2006;34:464–70. 20. Johanson M, Baer J, Hovermale H, Phouthavong P. Subtalar joint position during gastrocnemius stretching and ankle dorsifl exion range of motion. J Athl Train 2008;43:172–8.

Disabil Rehabil Downloaded from informahealthcare.com by Memorial University of Newfoundland on 08/01/14 For personal use only.

10

D. Cruz-Dı´az et al.

21. Reid A, Birmingham TB, Alcock G. Efficacy of mobilization with movement for patients with limited dorsiflexion after ankle sprain: a crossover trial. Physiother Can 2007;59:166–72. 22. Hertel J. Sensorimotor deficits with ankle sprains and chronic ankle instability. Clin Sports Med 2008;27:353–70. 23. van Rijn RM, van Os AG, Bernsen RMD, et al. What is the clinical course of acute ankle sprains? A systematic literature review. Am J Med 2008;121:324–31. 24. Mulligan BR. Mobilizations with movement. J Man Manip Ther 1993;1:154–6. 25. Riemann BL, Lephart SM. The sensorimotor system, part II: the role of proprioception in motor control and functional joint stability. J Athl Train 2002;37:80–4. 26. Bleakley CM, McDonough SM, MacAuley DC. Some conservative strategies are effective when added to controlled mobilisation with external support after acute ankle sprain: a systematic review. Aust J Physiother 2008;54:7–20. 27. Vicenzino B, Prangley I, Martin D. The initial effect of two Mulligan mobilizations with movement treatment techniques on ankle dorsiflexion. Australian Conference of Science and Medicine in Sport. A Sports Medicine Odyssey. Challenges, Controversies and Change [CD ROM]. Sports Medicine Australia; 2001. 28. Venturini C, Ituassu N, Teixeira L, Deus C. Intrarater and interrater reliability of two methods for measuring the active range of motion for ankle dorsifl exion in healthy subjects. Rev Bras Fisioter 2006; 10:407–11. 29. Carcia CR, Martin RL, Drouin JM. Validity of the Foot and Ankle Ability Measure in athletes with chronic ankle instability. J Athl Train 2008;43:179–83. 30. Cruz-Dı´az D, Hita-Contreras F, Lomas-Vega R, et al. Cross-cultural adaptation and validation of the Spanish version of the Cumberland Ankle Instability Tool (CAIT): an instrument to assess unilateral chronic ankle instability. Clin Rheumatol 2013;32:91–8. 31. Donahue M, Simon J, Docherty CL. Critical review of self-reported functional ankle instability measures. Foot Ankle Int/American Orthopaedic Foot and Ankle Society [and] Swiss Foot and Ankle Society 2011;32:1140–6.

Disabil Rehabil, Early Online: 1–10

32. Hiller CE, Refshauge KM, Bundy AC, et al. The Cumberland ankle instability tool: a report of validity and reliability testing. Arch Phys Med Rehabil 2006;87:1235–41. 33. Hoch MC, Staton GS, McKeon PO. Dorsiflexion range of motion significantly influences dynamic balance. J Sci Med Sport 2011;14: 90–2. 34. Robinson RH, Gribble PA. Support for a reduction in the number of trials needed for the Star Excursion Balance Test. Arch Phys Med Rehabil 2008;89:364–70. 35. Gribble PA, Hertel J. Considerations for normoralizing measures of the Star Excursion Balance Test. Meas Phys Ed Exer Sci 2003;7: 89–100. 36. van der Wees PJ, Lenssen AF, Hendriks EJ, et al. Effectiveness of exercise therapy and manual mobilisation in ankle sprain and functional instability: a systematic review. Aust J Physiother 2006; 52:27–37. 37. Collins N, Teys P, Vicenzino B. The initial effects of a Mulligan’s mobilization with movement technique on dorsiflexion and pain in subacute ankle sprains. Man Ther 2004;9:77–82. 38. Mattacola CG, Dwyer MK. Rehabilitation of the ankle after acute sprain or chronic instability. J Athl Train 2002;37:413–29. 39. Hoch MC, McKeon PO. The effectiveness of mobilization with movement at improving dorsiflexion after ankle sprain. J Sport Rehabil 2010;19:226–32. 40. Mangus BC, Hoffman LA, Hoffman MA, Altenburger P. Basic principles of extremity joint mobilization using a Kaltenborn approach. J Sport Rehabil 2002;11:235–50. 41. McKeon PO, Ingersoll CD, Kerrigan DC, et al. Balance training improves function and postural control in those with chronic ankle instability. Med Sci Sports Exerc 2008;40:1810–19. 42. Hoch MC, McKeon PO. Joint mobilization improves spatiotemporal postural control and range of motion in those with chronic ankle insta-bility. J Orthop Res 2011;29:326–32. 43. Hoch MC1, Staton GS, McKeon PO. Dorsiflexion range of motion significantly influences dynamic balance. J Sci Med Sport 2011;14: 90–2.