Article Text

Download PDFPDF

Therapeutic interventions for improving self-reported function in patients with chronic ankle instability: a systematic review
  1. Kyle B Kosik1,
  2. Ryan S McCann1,
  3. Masafumi Terada2,
  4. Phillip A Gribble1
  1. 1Department of Rehabilitation Sciences, University of Kentucky, Lexington, Kentucky, USA
  2. 2College of Sport and Health Sciences, Ritsumeikan University, Kyoto, Japan
  1. Correspondence to Kyle B Kosik, University of Kentucky, 214A Wethington Building 900 South Limestone Street, Lexington, KY 40536-0200, USA; Kyle.kosik{at}uky.edu

Abstract

Objective To identify which therapeutic intervention may be most effective for improving self-reported function in patients with chronic ankle instability (CAI).

Design Systematic literature review. Articles were appraised using the Downs and Black Checklist by 3 reviewers.

Data sources PubMed along with CINAHL, MEDLINE and SPORTDiscus within EBSCOhost for pertinent articles from their inception through August 2016.

Eligibility criteria for selected studies Articles included were required to (1) be written in English, (2) report adequate data to calculate effect sizes, (3) identify patients with CAI, (4) use some form of therapeutic intervention and (5) use a self-reported questionnaire as a main outcome measurement.

Results A broad spectrum of therapeutic interventions were identified related to balance training, multimodal rehabilitation, joint mobilisation, resistive training, soft-tissue mobilisation, passive calf stretching and orthotics. All of the articles included in the balance training category had moderate-to-strong Hedges g with none of the 95% CIs crossing 0. Hedges g effect sizes ranged from −0.67 to −2.31 and −0.51 to −1.43 for activities of daily living and physical activity, respectively. The multimodal rehabilitation category also produced moderate-to-strong Hedges g effect sizes but with large CIs crossing 0. Hedges g effect sizes ranged from −0.47 to −9.29 and −0.62 to −24.29 for activities of daily living and physical activity, respectively.

Conclusions The main findings from this systematic review were balance training provided the most consistent improvements in self-reported function for patients with CAI.

  • Ankle
  • Rehabilitation
  • Intervention effectiveness

Statistics from Altmetric.com

Introduction

Lateral ankle sprains have been documented as the most prevalent lower extremity musculoskeletal injury among physically active individuals.1 ,2 In the USA alone, it has been estimated that over two million ankle sprains occur each year.3 Despite being viewed as a musculoskeletal injury with minor consequences, a large portion of those with a previous ankle injury history will experience persistent pain, swelling and self-reported dysfunction4 and develop what is commonly referred as chronic ankle instability (CAI).5 Particularly, individuals with CAI report having recurrent injuries,6 perceived ankle instability and/or ‘giving-way’,7 and decreased self-reported function.8 The development of CAI is also characterised by structural alterations,9 ,10 sensorimotor impairments,11–13 altered movement patterns14–16 and participation restrictions.17 These changes often result in a reduction in the overall health-related quality of life18 and an increased risk of other chronic diseases.19

The majority of available research has focused on identifying the disease-oriented factors that contribute to CAI. Specifically, researchers have suggested that both mechanical and sensorimotor insufficiencies may be present in those with CAI.5 Identified mechanical insufficiencies include arthrokinematic restrictions,20 ,21 ligamentous laxity22–24 and joint degeneration,25 whereas sensorimotor deficiencies such as diminished static26–28 and dynamic postural control,29–31 loss of strength32 ,33 and altered spinal reflex excitability34–36 have also been found in patients with CAI. This has led to the development of several therapeutic interventions such as balance training,37–39 resistive training,40 ,41 joint mobilisation42 ,43 and multimodal rehabilitations44 ,45 programmes that have all led to the improvements in the mechanical and sensorimotor insufficiencies previously mentioned.

Patient-oriented evidence has been increasingly recognised in healthcare as an important component of evidence-based medicine because it provides valuable information on how the patient perceives their health condition.46 Patient-oriented evidence is most commonly derived through the use of self-reported questionnaires in patients with CAI to evaluate the functional limitations they experience.47 Evidence has shown that patients with CAI regularly report having greater functional limitations compared with their healthy counterparts.48 Further, the sensorimotor and mechanical insufficiencies associated with CAI also contribute to the increased self-reported dysfunction linked to CAI.18 ,49 ,50 Therefore, researchers48 ,51 and the International Ankle Consortium52–54 have advocated for the use of appropriate region-specific self-reported measures to determine the level of function and for evaluating the effectiveness of a therapeutic intervention for patients with CAI.55 This has led to an increase in reporting of self-reported measures of function to assess the efficacy of a variety of therapeutic interventions.56–58 However, inconsistent results exist for whether therapeutic interventions can improve self-reported measures of function in patients with CAI.57 ,59

The diverse treatment options, the significance of patient-oriented evidence and the inconsistent results for improving self-reported function warrants a comprehensive review of the literature to identify the most effective therapeutic intervention for improving self-reported function in patients with CAI. Identifying the most effective therapeutic intervention for improving self-reported function will guide future investigations in designing rehabilitation and direct clinicians in the exercises they prescribe to patients. Therefore, the purpose of this systematic review is to identify which therapeutic intervention is most effective at improving self-reported function in patients with CAI.

Methods

Literature search strategy

We followed the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines.60 The primary author performed comprehensive electronic literature search in August 2016 using PubMed along with CINAHL with full text; MEDLINE and SPORTDiscus were searched within the EBSCOhost database (table 1, see online supplementary file) for peer-reviewed articles. A mix of key words was used to identify relevant articles:

  • P: chronic ankle instability OR mechanical ankle instability OR functional ankle instability OR ankle instability.

  • I: rehabilitation OR balance training OR strength training OR joint mobilization OR manual therapy OR wobble board OR proprioceptive neuromuscular facilitation OR plyometric training OR orthotics OR therapeutic ultrasound OR transcranial electrical neuromuscular facilitation OR cryotherapy OR imagery OR non-steroidal anti-inflammatory drugs OR complementary therapy OR alternative therapy OR hyperbaric oxygen therapy.

  • C: none.

  • O: self-reported function OR patient generated outcome OR perceived function OR Functional Ankle Disability Index OR Cumberland Ankle Instability Tool OR Foot and Ankle Ability Measure OR Chronic ankle instability scale OR Ankle Joint Functional Assessment Tool.

Table 1

Risk of bias

supplementary data

Systematic search strategy with number of studies identified for each term(s).

We included studies published since 1965 that were written in English and used human participants. The primary author solely searched each database systematically and selected potential articles based on title and abstract alone. The primary author, along with two independent reviewers, then carefully reviewed each of the potential articles using the inclusion criteria to identify the specific articles that answered our research question. In addition to the electronic search, a hand search of reference lists of the articles screened for inclusion was performed to identify any additional articles.

Selection criteria

We included peer-reviewed articles if they met the following inclusion criteria:

  1. Articles with a level 4 evidence or higher were included because we wanted to provide researchers with a comprehensive literature search of all of the possible treatment options.

  2. Participants identified as having CAI, functional ankle instability, ankle instability or mechanical instability were included and received an intervention to improve self-reported function.

  3. The authors had to use a region-specific Patient Reported Outcomes questionnaire as an outcome measurement that provides either self-reported (1) function, (2) instability or (3) disability.

  4. The authors were not required to compare the intervention with a placebo, control or other standard of care. However, the intervention was required to be performed more than once, as this better reflected common clinical practice.

  5. Authors reported means and SDs of pretreatment and post-treatment, or baseline and follow-up measures for calculating a standardised effect size. If these data were not available, the corresponding author of the research article was contacted to request these data before excluding the article.

Data extraction

The literature search identified several therapeutic interventions. Articles were first filtered into categories that best described the therapeutic intervention. If the article compared two separate therapeutic interventions, the groups were separated and included into the appropriate category. Based on the information from the articles included, the categories were: (1) balance training, (2) joint mobilisation, (3) multimodal, (4) resistive training, (5) soft-tissue mobilisation, (6) orthotics and (7) static stretching. Furthermore, many of the peer-reviewed articles investigated the effect of the selected interventions on self-reported function of the ankle during activities of daily living and physical activity. Data were further subdivided into categories representing self-reported function during activities of daily living or physical activity.

The primary author extracted all of the necessary information for the systematic review and calculated the Hedges g. Data extracted from the studies included means, SDs, sample size, inclusion criteria, demographic information, form of intervention and the duration/frequency of the intervention. If the necessary data were not available, the corresponding author of each included study was contacted for the information. Studies were excluded if the corresponding author did not respond or provide adequate information.

Risk of bias assessment

Research articles that met the inclusion criteria were evaluated for their risk of bias by the primary author and two independent reviewers using the Downs and Black Checklist. The Downs and Black Checklist was developed to evaluate the risk of bias for non-randomised and randomised control trials.61 The checklist is comprised of 27 questions designed to evaluate reporting and selection bias along with external and internal validity of an article. A higher score on the Downs and Black Checklist is associated with higher quality evidence. If all three co-authors did not agree on the study's score, the authors discussed among themselves until they came to a mutual agreement on a score. Following the critical appraisal, each article was given a level of evidence in accordance with the Oxford Centre for Evidence-Based Medicine (CEBM) guidelines.62 The levels of evidence range from 1 to 5, where level 1 evidence represents the highest quality and level 5 is the lowest quality.62

Strength of recommendation

The Strength of Recommendation of Taxonomy (SORT) is a scale that provides clinicians and researchers with a grade of recommendation for clinical practice based on the overall current body of evidence.63 The strength of recommendation takes into consideration the level of evidence, the number of articles, and the consistency and coherence of the evidence in its entirety. The SORT scale includes ratings of A, B or C.62 An ‘A’ is given when there is consistent and good-quality patient-oriented evidence, ‘B’ represents inconsistent or limited-quality patient-oriented evidence, and ‘C’ if evidence is based on studies of diagnosis, expert opinions, disease-oriented outcomes or case series.62 A grade of recommendation was given to each category based on the articles included into the final analysis. The primary author and the two independent reviewers responsible for grading the risk of bias of each article were also responsible for discussing and arrive at a mutual agreement on a grade of recommendation for each category.

Statistical analysis

A meta-analysis was not performed because of the multiple therapeutic interventions identified. Rather, we performed a qualitative analysis in order to provide a comprehensive overview of all possible treatments for improving self-reported function and provide recommendations based on the best available evidence.

Hedges g effect sizes were calculated to examine the magnitude of the difference between the pretreatment and post-treatment scores or baseline and follow-up data for each study. If effect sizes were provided in the original publication, we recalculated the effect size for consistency by comparing across all the studies included in this review. The strength of the effect size was interpreted as weak (d<0.4), moderate (0.4≤d<0.8) and strong (d≥0.8).64 Based on the self-reported questionnaires identified and included in this review, a negative effect size indicates improved self-reported function following the completion of the intervention. Additionally, we calculated the 95% CIs around the effect sizes. Both the magnitude of the effect size and the width of the CI should be considered when interpreting the magnitude of the treatment effect. For example, when a large treatment effect size is present with a wide CI crossing zero, it cannot be concluded if a truly beneficial or harmful effect would occur. Conversely, a large treatment effect size with narrow CIs not crossing zero would indicate a truly beneficial or harmful effect and would provide strong clinically meaningful information.

Results

The initial search strategy retrieved 167 articles (figure 1). Twenty-eight articles were identified based on title and abstract alone. Following careful review, 20 articles met the inclusion criteria. Three of 20 articles were excluded from the current review because of insufficient information for data analysis.44 ,45 ,65 A total of 17 articles were included into the final analysis. A summary of the inclusion criteria, sample size, length and frequency of the intervention, form of measurement and results are listed in table 2 (see online supplementary file). Researchers provided self-reported function for activities of daily living in 16 articles,51 ,56–59 ,66–75 whereas only 13 articles provided self-reported function during physical activity.43 ,51 ,57 ,59 ,68 ,70–72 ,74–77 Seventeen articles were classified into the following categories based on the description of the therapeutic intervention: 6 included balance training,66 ,68–71 ,77 6 included a multimodal rehabilitation protocol51 ,68 ,72–75 and 5 included joint mobilisations,43 ,57–59 ,76 of which 2 included resistive training,69 ,71 1 included soft-tissue mobilisation,57 1 included calf-stretching57 and 1 included orthotics.56

Table 2

Effect sizes and 95% CIs for activities of daily living

Figure 1

Flow chart for selected research articles included for the systematic review.

supplementary data

Summary of articles included

Risk of bias

A summary of the study design and scores on the Downs and Black Checklist are listed in table 1. The quality of the studies based on the average Downs and Black Checklist was 19, with a range of 11–24.74 ,76 Study designs from these studies included randomised control trial, cohort study, case–control and case series. Subsequently, the level of evidence ranged from 1b to 4 (table 1).

A summary of the PRO scores, effect sizes and associated 95% CIs during activities of daily living and physical activity is provided in tables 2 and 3.

Table 3

Effect sizes and 95% CIs for physical activity

A summary of the strength of recommendation for each category is provided in table 4.

Table 4

Strength of Recommendation Taxonomy (SORT)

Balance training

Two articles assessed the efficacy of balance training during activities of daily living only,66 ,69 while four research articles evaluated the efficacy of balance training for activities of daily living and physical activity.68 ,70 ,71 ,77 Four articles were randomised control trials68 ,69 ,71 ,77 and two were case–control studies.66 ,70 The average risk of bias of the six articles was 19 with a range of 16–2266 ,71 on the Downs and Black Checklist (table 1). Based on the number of articles included, the overall high quality of evidence and consistent findings, we provided an ‘A’ for the strength of recommendation for activities of daily living scale and physical activity subscale (table 4).

Activities of daily living scale

Treatment effect sizes were considered moderate to strong as they ranged from −0.6770 to −2.31.66 More specifically, strong effect sizes were observed for the Foot and Ankle Ability Measure (FAAM; Hedges g: −0.9371 to −1.4468), Cumberland Ankle Instability Tool (CAIT; Hedges g: −1.1571 to −2.1069) and Ankle Joint Functional Assessment Tool (AJFAT) (Hedges g: −2.3166); whereas moderate-to-strong effect sizes were observed for the Foot and Ankle Disability Index (FADI; Hedges g: −0.6770 to −1.0177). Further, none of the 95% CIs included zero (table 4), suggesting that balance training is beneficial for improving self-reported measurements related to activities of daily living.

Physical activity subscale

Treatment effect sizes were considered moderate to strong as they ranged from −0.5170 to −1.1177 for the FADI and −0.9471 to −1.4368 on the FAAM, with none of the CIs crossing zero. These results indicate that balance training provided favourable outcomes for improving self-reported function during physical activity.68 ,70 ,71 ,77

Multimodal

The majority of these research articles examined the combined effects of an intervention coupled with a standard of care treatment. Five articles51 ,68 ,72 ,74 ,75 included in this category examined the effects of a multimodal rehabilitation intervention for improving self-reported function during activities of daily living and physical activity, while one study only assessed activities of daily living.73 There were five articles identified as randomised control clinical trials,51 ,68 ,73–75 and one case–control study.72 The risk of bias of these studies had an average of 18.3 with a range of 15–2372 ,73 on the Downs and Black Checklist (table 1). Based on the number of articles included and inconsistent results, we provided a ‘B’ for the strength of recommendation for both categories (table 4).

Activities of daily living scale

Treatment effect sizes ranged from −9.2972 to −0.4774 (table 2). The range of treatment effect sizes for the FAAM was −0.4774 to −1.6768 and −1.3773 to −9.2972 for the FADI. However, two articles did have CIs that included zero, implying that a truly beneficial treatment effect cannot be concluded.74 ,75 Collectively, these results suggest that supervised rehabilitation programmes are effective at improving self-reported function.

Physical activity subscale

Treatment effect sizes ranged from −24.2972 to −0.4574 (table 3), with two articles having CIs crossing zero.74 ,75 Overall, these results suggest that multimodal rehabilitation is effective at improving self-reported function during physical activity.

Joint mobilisation

Three peer-reviewed articles examined the efficacy of joint mobilisation alone on self-reported function during activities of daily living and physical activity.43 ,57 ,59 Two studies examined the efficacy of joint mobilisation on either self-reported function during activities of daily living58 or physical activity only.76 Three randomised control clinical trials57 ,58 ,76 and two case series43 ,59 were identified among these studies. The risk of bias of these articles had an average Downs and Black Checklist score of 21 with a range of 18–2443 ,76 (table 3). Based on the number of lower quality of evidence and inconsistent findings, we provided a grade of ‘C’ for the strength of recommendation for both categories (table 4).

Activities of daily living scale

The treatment effect sizes were considered weak to strong and ranged from −0.3459 to −5.41.58 The effect sizes were weak to strong for the FAAM (Hedges g ranged −0.3459 to −0.7343) and the CAIT (Hedges g: −0.1958 to −5.4158). However, only one article had CIs that did not cross zero.58 Therefore, it cannot be determined whether a truly beneficial or harmful treatment effect exists.

Physical activity subscale

Despite the strong effect sizes (Hedges g: −0.9843 to −0.8457) observed in two articles, effect sizes ranged from weak to moderate (Hedges g: −0.2376 to −0.5759) to harmful (Hedges g: 0.0376) with CIs crossing zero.

Resistive training

Two articles examined the efficacy of resistive training alone for improving self-reported function during activities of daily living,69 ,71 while only one of the studies focused on improving self-reported function during physical activity.71 Both articles were randomised control trials with an average score of 21.5 ranging from 2169 to 2271 on the Downs and Black Checklist (table 3). Considering the few articles retrieved with high quality of evidence and consistent results, we provided a ‘B’ for the strength of recommendation (table 4).

Activities of daily living scale

Treatment effect sizes were considered moderate to weak for the CAIT (Hedges g: −0.5971 to −0.1169) and weak for the FAAM (Hedges g: −0.2571), with CI crossing zero. These results suggest that resistive training does not provide clinically meaningful improvement in self-reported function.

Physical activity subscale

Only one article78 examined the effect of resistive training on self-reported function during physical activity and found a moderate effect size (Hedges g: −0.55) on the FAAM-Sport with CIs crossing zero, suggesting resistive training does not provide clinically meaningful improvements.

Stretching

The only study57 included in this category had a strong effect size on the FAAM (Hedges g: −0.72) during activities of daily living with the CIs not crossing zero (table 2). In contrast, a moderate effect size was observed on the FAAM-Sport (Hedges g: −0.40) with CIs crossing zero. These results suggest that static stretching of the triceps surae can improve self-reported function during activities of daily living but not physical activity. Furthermore, considering that only one high-quality article was identified, we provided a grade ‘B’ for the strength of recommendation (table 4).

Soft-tissue mobilisation

Only one study57 was included in this category, with a moderate effect size for the FAAM (Hedges g: −0.61) and FAAM-Sport (Hedges g: −0.45) with CIs that crossed zero for each questionnaire. Therefore, these results imply that six sessions of plantar massage does not provide clinically meaningful improvements on the FAAM or FAAM-Sport. Based on only one high-quality article identified, we provided a grade ‘B’ for the strength of recommendation (table 4).

Orthotics

Only one study56 met the inclusion criteria for this category, with a moderate treatment effect (Hedges g: −0.46) on the CAIT with CIs crossing zero, suggesting that orthotics do not provide a clinically meaningful improvement in self-reported function. Finally, because of the single article with a case-series study design, we provided a grade ‘C’ for the strength of recommendation (table 4).

Discussion

After reviewing the literature, we found that balance training and multimodal rehabilitation programmes are both beneficial, but balance training appears to have the greatest impact on self-reported function in patients with CAI.

Balance training

There is convincing evidence to suggest that balance training improves self-reported function in those with CAI as strong-to-moderate effect sizes and narrow CIs were found (tables 2 and 3). The presence of CAI places constraints on the sensorimotor system that can lead to altered mechanics during functional tasks,5 ,11 translating to decreased levels of self-reported function.48 While there is preliminary evidence to support the use of balance training to decrease the subjective feeling of ‘giving-way’ at the ankle,79 the current review provides additional support for the use of balance training to improve subjective functional limitations for those with CAI.

As expected, there was a large amount of variability within the balance training programmes, including the type of exercises, the rate of progression and length of the intervention. Despite this heterogeneity, they all progressively increased the demands placed on the sensorimotor system by changing the task and/or the environment. For example, changes within the task were carried out by restricting the ability of the participants to use their hands, closing their eyes and/or introducing unanticipated movements.69 ,77 To further increase the demands placed on the sensorimotor system, authors purposefully manipulated the environment through the introduction of unstable surfaces.69 Therefore, by constantly challenging the sensorimotor system to respond efficiently to changes in the task and/or environment, participants likely developed new movement strategies to overcome their functional limitations experienced during activities of daily living and physical activity77 which lead to an increase in self-reported function.

In addition to the data presented, there is evidence to suggest that balance training is effective at reducing the risk of injury.80 Verhagen et al80 prospectively demonstrated balance training provides even greater benefit at reducing the risk of injury in those with a previous ankle joint injury. Furthermore, balance training is an effective intervention for improving postural control deficits in those with CAI. Most of the studies included in this review also examined the effect of balance training for improving static and dynamic postural control deficits and found positive results.66 ,77

Therefore, based on the available evidence for balance training to improve subjective and objective measures of function and reduce the risk of reinjury, researchers and clinicians should implement balance training into their future rehabilitation programmes, specifically, a balance training programme that progressively challenges the sensorimotor system through purposeful manipulation of the task and environment to foster the development of new movement strategies.

Multimodal

There is moderate evidence that multimodal rehabilitation programmes improve self-reported function in those with CAI. All of the effect sizes ranged from strong to moderate for the activities of daily living and physical activity subscale; however, the CIs crossed the zero line for two articles, indicating that a truly beneficial treatment effect cannot be concluded (tables 2 and 3).74 ,75 Lee et al72 observed the largest treatment effect on the FADI and the FADI-Sport when they performed a high-intensity supervised training programme focused on strength, flexibility, balance and agility. On further inspection of the baseline self-reported questionnaire scores, it appears the participants enrolled were extremely low functioning with an average score of below 40% on the FADI. Therefore, the large treatment effect size may be attributed to the low-functioning individuals enrolled, as they likely had greater room for improvements in self-reported function.

The multiple sensorimotor and mechanical insufficiencies exhibited in patients with CAI5 and that are associated with self-reported function49 ,50 make multimodal rehabilitation strategies popular among clinicians. The multimodal rehabilitation programmes included in this review were general rehabilitation programmes that were largely augmented with a novel therapeutic exercise such as Graston instrument-assisted soft-tissue mobilisation68 or trigger-point dry needling.75 Donovan and Hertel81 recently introduced a paradigm that emphasised the need for assessing and conservatively treating the unique deficits seen by individual patients with CAI. Specifically, the paradigm encourages clinicians to assess for restrictions in range of motion, loss of strength, decreased balance and altered movement patterns during walking, running and jumping.81 Once the deficits are identified, the authors recommend treating them with targeted exercises and reassessing for improvements.81 Though the majority of the studies included in our review likely did not use this framework for constructing their rehabilitation programme, all of the multimodal programmes used therapeutic exercises with some combination of strength training, range of motion and balance. Therefore, because these intervention programmes focused on three of the primary domains described and emphasised by Dovovan and Hertel,81 they likely addressed a combination of the mechanical and sensorimotor insufficiencies observed in individual patients with CAI, thus, explaining the increase in self-reported function found in the current review.

However, when considering the time needed to perform a multimodal rehabilitation programme and the limited time clinicians often have with patients it is important to identify which combination of exercises maximises outcomes. The current review found strong evidence for the use of a progressive balance training programme when used in isolation for targeting self-reported function, but limited evidence for the use of strength training and joint mobilisation when performed alone. Therefore, it would appear that balance training might be the most important factor to include in a rehabilitation programme to target self-reported function. Whereas, the inclusion of other exercises including strength and joint mobilisation, might be more suited for targeting objective measures such as strength deficits82 or range of motion restrictions when found within specific patients, self-reported function is not. More research is needed to identify which combination of exercises produces greatest improvements in self-reported function in patients with CAI.

Joint mobilisation

Limited supportive evidence was found for the use of joint mobilisation in isolation for improving self-reported function during activities of daily living for patients with CAI. Only one article demonstrated favourable outcomes following six sessions of weight-bearing talocrural joint mobilisation with movement immediately following the intervention.58 Conversely, Hoch et al43 and McKeon and Wikstrom57 also performed six sessions of Maitland grade III anterior-to-posterior talocrual joint mobilisations and did not find a truly beneficial treatment effect immediately following the intervention (table 2). The prolonged symptoms associated with CAI potentially may arise in part from dorsiflexion restrictions.83–85 Normal dorsiflexion range of motion requires the talus to roll upwards and glide posteriorly with the fibula simultaneously moving superiorly and lateral.86 However, patients with CAI have been found to have an anteriorly displaced talus that can prevent the talus from gliding posteriorly and achieving full dorsiflexion range of motion.10 Anterior-to-posterior talocrural joint mobilisation has been purported as a therapeutic intervention that can re-establish normal arthrokinematic motion in patients with CAI.42 ,43 ,58 However, when considering the paradigm by Donovan and Hertel,81 not everyone with CAI may present with decreased dorsiflexion range of motion which may explain the inconsistency found in the current review as Cruz-Diaz et al58 were the only ones to assess for restrictions in dorsiflexion range of motion prior to enrolment.

Although they did not find improvements on the FAAM-ADL, Hoch et al43 and Mckeon et al77 were the only ones to a truly beneficial effect for the FAAM-Sport immediately after the intervention. The lack of consistency for improvements in self-reported function during physical activity and not activities of daily living may be attributed to association found previously between dorsiflexion range of motion and the FAAM-Sport subscale.18 Based on the available evidence, there is mixed evidence that joint mobilisation in isolation improves self-reported function. Joint mobilisations may be more effective at restoring objective measures such as range of motion or dynamic postural control and should be used in combination with other therapeutic interventions.

Strength training

Our review yielded limited evidence for the efficacy of strength training protocols for improving self-reported function, with only two articles retrieved.69 ,71 Both articles received high marks on the methodological rating scales that produced small-to-moderate effect sizes with large CI crossing zero. Strength deficits of the main ankle stabilisers have been thought to predispose individuals to weakness and recurrent injuries. Early reports provided support by demonstrating strength deficits in the surrounding ankle musculature,32 ,87 ,88 though some researchers have failed to identify strength deficits in patients with CAI.24 ,89 This inconsistency among the literature suggests that strength deficits may not be an important factor in the development of self-reported function in patients with CAI. Rather, timing of muscle activation during functional tasks such as walking,90 jumping91 and balancing92 may be a more important factor in self-reported function. Though strength training programmes have found to improve measures of strength,82 based on the high-quality articles included into this category demonstrating poor-to-moderate effect sizes with large CIs, it would appear that resistive training does not lead to clinically meaningful improvements in self-reported function.

Orthotics

We only found one article56 that examined the effect orthotics have on self-reported function and found inconclusive results (table 2). Orthotics are thought to improve mechanical stability of the subtalar joint and provide sensory feedback. There is a large body of evidence to warrant the use of orthotics for other objective measures of function in people with CAI.93 ,94 However, it is difficult to make any definitive clinical conclusion for the use of orthotics for improving self-reported function in patients with CAI based on one study that was limited in its methodological quality. More high-quality evidence is needed to further substantiate or refute the current findings.

Static stretching and soft-tissue mobilisation

McKeon and Wikstrom57 used static stretching and soft-tissue mobilisation to target the sensory information arising from the foot and ankle that is associated with the presence of CAI95–98 and self-reported function.96 Only static stretching found positive improvements for improving self-reported function during activities of daily living (table 2). Interestingly, the authors also used joint mobilisations to target the sensory alterations arising from the ankle joint and found improvements in physical activity but not self-reported function during activities of daily living (tables 2 and 3). Both joint mobilisation and static stretching appear to be beneficial for improving objective measures such as dorsiflexion range of motion,57 but the relationship between joint mobilisation and static stretching for improving self-reported function is unclear. The high methodological rating score and clinically meaningful improvements in self-reported function McKeon and Wikstrom57 warrants the need for more research on static stretching and the use other sensory targeted interventions to improve self-reported function.

Limitations

While we followed PRISMA guidelines, this systematic review is not without limitations. First, we limited our search strategy to include articles that were written only in English, only a single investigator screened for potential articles based on the title and abstract for inclusion and the lack of duplicate data extraction. Second, we did not restrict our analysis to any particular study design or level of evidence. For example, a variety of research designs with a broad range of methodological scores were included in this review. Rather than excluding these articles, we included any article that met our inclusion/exclusion criteria because we wanted to provide a comprehensive review of the literature to identify all potential therapeutic interventions for improving self-reported function. In doing so, we hoped to provide a direction for future researchers and clinicians in designing more effective rehabilitation protocols through more rigorous research designs. Moreover, most of the participants in the studies that were included in this review were not blinded to their therapeutic intervention, which may have influenced the improvements seen in this review. Finally, we did not perform a meta-analysis because of the wide variety of therapeutic interventions, heterogeneity of methodological designs and limited evidence for most interventions.

Conclusions

Evaluating self-reported function with PROs is an important criterion for understanding how the patient perceives their health condition and for evaluating the efficacy of rehabilitation protocols. We identified a wide variety of other therapeutic interventions within the literature to improve self-reported function; however, balance training and multimodal rehabilitation programmes appear to effectively improve self-reported function. Furthermore, the data presented in this systematic review indicate that progressive balance training was the most consistent form of therapeutic intervention. Clinicians should implement a progressive balance training programme that is focused on continuously challenging the sensorimotor system by increasing the difficulty of the task or by changing the environment in which the tasks are being performed.

What are the findings?

  • Chronic ankle instability (CAI) is commonly associated with decreased self-reported function.

  • A variety of therapeutic interventions have been used to improve self-reported function for patients with CAI.

  • It is unknown which therapeutic intervention is the most effective at improving self-reported function in patients with CAI.

How might it impact on clinical practice in the future?

  • Balance training is the most consistent therapeutic intervention for improving self-reported function in patients with CAI.

  • Majority of the multimodal rehabilitation programmes provided favourable outcomes; however, further evidence is needed to verify these findings.

  • Limited evidence was found for the use of other therapeutic interventions targeting strength deficits and sensory alterations associated with CAI.

References

Footnotes

  • Twitter Follow Phillip Gribble at @gribblepa

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles