Abstract
Objectives
Sensory Targeted Ankle Rehabilitation Strategies that stimulate sensory receptors improve postural control in chronic ankle instability participants. However, not all participants have equal responses. Therefore, identifying predictors of treatment success is needed to improve clinician efficiency when treating chronic ankle instability. Therefore, the purpose was to identify predictors of successfully improving postural control in chronic ankle instability participants.
Design
Secondary data analysis
Methods
Fifty-nine participants with self-reported chronic ankle instability participated. The condition was defined as a history of at least two episodes of “giving way” within the past 6 months; and limitations in self-reported function as measured by the Foot and Ankle Ability Measure. Participants were randomized into three treatment groups (plantar massage, ankle joint mobilization, calf stretching) that received 6, 5-minute treatment sessions over a 2-week period. The main outcome measure was treatment success, defined as a participant exceeding the minimal detectable change score for a clinician-oriented single limb balance test.
Results
Participants with ≥3 balance test errors had a 73% probability of treatment success following ankle joint mobilizations. Participants with a self-reported function between limb difference <16.07% and who made >2.5 errors had a 99% probability of treatment success following plantar massage. Those who sustained ≥11 ankle sprains had a 94% treatment success probability following calf stretching.
Conclusions
Self-reported functional deficits, worse single limb balance, and number of previous ankle sprains are important characteristics when determining if chronic ankle instability participants will have an increased probability of treatment success.
Keywords: Massage, Musculoskeletal Manipulations, Joint Mobilization, Static Stretching, joint instability
Lateral ankle sprains are the most common sports related injuries.1 Epidemiological data illustrate that over 625,000 ankle sprains reported to United States emergency departments each year between 2002–2006.2 Unfortunately, this data does not capture the ankle sprains treated by other allied health care providers or the individuals who do not seek treatment3, making the true incidence difficult to determine. However, at least 33% of those who sustain a lateral ankle sprain develop CAI4 with similar and sometimes higher incidence rates being observed in sporting populations5 as well as in children.6 Chronic ankle instability is characterized by two hallmark symptoms: repeated episodes of giving-way and recurrent sprains with or without the presence of a mechanical instability of the joint.7 Those with CAI have also been reported to have a number of structural and/or sensorimotor symptoms,8 lowered quality of life scores,9 and decreased physical activity levels.10 A link also exists between CAI and post-traumatic ankle osteoarthritis, with 68–78% of CAI participants developing ankle osteoarthritis.11 One of the most commonly examined sensorimotor outcome measures is single leg balance because impaired balance is associated with an increased risk of ankle injury.12
Evidence clearly demonstrates that balance training programs for those with CAI, not only improve postural control13 but also result in a reduction of recurrent ankle sprains.14. As a result, balance training is frequently used to treat acute ankle sprains and those with CAI. However, research has started exploring the possibility of improving balance by stimulating sensory receptors around the foot/ankle complex.15–17 Cumulative findings indicate that balance can be improved following the stimulation of sensory receptors. Most recently, a randomized controlled trial (RCT) found that those who received 6, 5-minute treatments of either plantar massage, joint mobilization, or calf stretching over a 2-week period of time had significant improvements in ROM, postural control, and self-reported function.18
Unfortunately, none of the investigations, including our RCT, assigned interventions based on specific deficits associated with the individual participant. Therefore, it is possible that some participants could have experienced a more dramatic improvement in single-limb balance. If true, then other individuals who portray similar characteristics might also experience a dramatic improvement and thus should be targeted to undergo one of these interventions. Clinical prediction rules (CPRs) have the potential to provide clinicians with such data by assisting them in the identification of patient subgroups.19, 20 While CPRs have been developed for treating non-specific low back pain21, patellofemoral pain22, and acute ankle sprains with manipulative therapy23, no CPR is available for treating those with CAI. The use of CPRs in those with CAI, would provide specific evidence in support of the Impairment-Based rehabilitation model proposed by Donovan & Hertel24 and could help guide clinical decision making with regards to treating balance deficits. Therefore the purpose of this secondary analysis18 was to identify possible predictors of successfully improving a clinical measure of balance in those with CAI using joint mobilizations, plantar massage, or calf stretching. More specifically, participant characteristics (e.g. age, height, weight, etc), injury history (e.g. number of sprains, giving way episodes, etc) and outcomes captured during a baseline assessment were evaluated to determine if they could predict which participants would have meaningful improvements in a clinician-oriented measure of balance following a 2-week intervention.
Methods
As indicated, this is a secondary analysis of data from a multicenter, mutli-arm parallel randomized control phase II clinical trial (Trial # NCT01541657) with a 1-month follow-up period.18 This trial, tested the effectiveness of ankle joint mobilization, plantar massage, and triceps surae stretching relative to a control condition at improving patient-, clinician-, and laboratory-based outcome measures in individuals with CAI. The protocol was approved by the research and ethics committees for each of the large public universities where the RCT was conducted and all participants provided written informed consent prior to participation. For this analysis, only the treatment groups were used.
As described in the original RCT report,18 CAI participants were recruited from the general population (i.e. student, staff, and faculty) of three large public universities in the United States. These participants were physical active young adults and while this investigation was initiated prior to the International Ankle Consortium’s published recommendations,25 the inclusion/exclusion criteria are consistent with those recommended. More specifically, those with CAI were required to have sustained at least two episodes of “giving way” within the past 6 months; score ≥5 on the Ankle Instability Instrument (AII), score < 90% on the FAAM, and score < 80% on the FAAM Sport (FAAM-S). Those with bilateral CAI were allowed to participate and the worse limb, based on FAAM & FAAM-S, then underwent the intervention. Exclusion criteria required participants to be free from: an acute ankle sprain in the 6 weeks prior to screening, a previous history of ankle surgeries, lower extremity surgeries associated with internal derangements or repairs, and/or other conditions known to affect sensorimotor function. Participant characteristics, based on their assigned treatment groups and treatment success, can be seen in Table 1.
Table 1.
Univariate comparisons between the Success and Unsuccessful groups based on SLBT scores exceeding the SLBT MDC sorted by treatment group.
| Ankle Joint Mobilization | Plantar Massage | Stretching | |||||||
|---|---|---|---|---|---|---|---|---|---|
|
| |||||||||
| Variable | Success (n=9, 45%) | Unsuccessful (n=11, 55%) | Significance | Success (n=10, 53%) | Unsuccessful (n=9, 47%) | Significance | Success (n=4, 25%) | Unsuccessful (n=16, 75%) | Significance |
| Sex: female, n (%) | 5 (55%) | 6 (55%) | 0.655 | 7 (70%) | 4 (44%) | 0.491 | 0 (0%) | 12 (75%) | 0.371 |
| Age (years) | 22.67±5.45 | 24.27±7.7 | 0.606 | 22.20±2.29 | 22.33±3.42 | 0.921 | 20.75±4.19 | 22.25±2.40 | 0.348 |
| Height (cm) | 171.32±10.32 | 172.26±9.42 | 0.832 | 170.81±5.49 | 174.06±10.13 | 0.391 | 181.61±8.85 | 168.11±12.82 | 0.064 a |
| Weight (cm) | 78.43±22.31 | 76.65±16.23 | 0.839 | 73.99±10.92 | 77.07±17.63 | 0.649 | 83.29±16.77 | 65.68±14.93 | 0.054 a |
| AII (# Yes responses) | 7.22±1.64 | 6.72±1.67 | 0.516 | 6.10±1.10 | 6.88±1.16 | 0.148 | 7.00±1.41 | 6.68±1.40 | 0.695 |
| # of Recurrent Sprains | 4.22±4.23 | 4.45±2.76 | 0.884 | 4.60±3.65 | 3.77±2.53 | 0.581 | 9.25±5.12 | 5.5±3.16 | 0.076 a |
| Time since last sprain (months) | 10.75±8.31 | 11.63±9.6 | 0.831 | 23.70±23.59 | 20.94±38.06 | 0.850 | 8.00±7.34 | 13.87±9.86 | 0.283 |
| # Giving way episodes within 6 months | 5.22±3.92 | 4.27±3.60 | 0.581 | 4.30±2.31 | 6.55±6.46 | 0.315 | 4.50±3.00 | 6.200±5.75 | 0.582 |
| FAAM (%) | 77.91±9.14 | 82.90±11.43 | 0.303 | 79.52±13.32 | 73.67±12.35 | 0.337 | 77.08±13.69 | 75.29±12.57 | 0.805 |
| FAAM between limb difference (%) | 15.74±12.18 | 8.08±5.48 | 0.075 a | 6.78±5.61 | 19.97±12.38 | 0.007 a | 18.88±5.46 | 29.45±10.47 | 0.070 a |
| FAAM-S (%) | 60.41±15.14 | 65.05±9.56 | 0.415 | 69.03±9.13 | 55.55±14.63 | 0.026 a | 51.56±20.65 | 63.39±12.86 | 0.160 |
| FAAM-S between limb difference (%) | 24.64±19.54 | 19.60±15.25 | 0.524 | 16.56±12.76 | 34.37±19.76 | 0.031 a | 16.40±19.98 | 25.66±14.88 | 0.310 |
| WBLT (cm) | 8.11±1.77 | 8.23±3.59 | 0.933 | 10.04±4.15 | 10.92±4.95 | 0.680 | 8.91±2.31 | 10.14±2.92 | 0.448 |
| WBLT between limb difference (CM) | 2.36±2.14 | 0.96±2.4 | 0.191 | 0.82±1.55 | 1.67±2.12 | 0.335 | 1.33±1.90 | 1.51±2.15 | 0.877 |
| SLBT (errors) | 4.03±1.95 | 1.84±1.72 | 0.016 a | 3.99±1.18 | 2.48±1.96 | 0.055 a | 4.25±1.52 | 2.79±2.71 | 0.321 |
| SLBT between limb difference (errors) | −0.44±1.94 | −0.03±1.55 | 0.602 | −1.60±1.47 | −0.41±0.79 | 0.045 a | −0.83±1.00 | −0.56±1.43 | 0.728 |
Indicates a significant difference (p≤0.10) between the success and unsuccessful groups for the listed intervention.
A total of 6, 5-minute treatment sessions of each participant’s randomly assigned intervention were applied over a two-week period. Treatments were distributed over the two week period (e.g. 3 per week with at least 48 hours between treatments) as best as possible given participant availability. Ankle joint mobilizations consisted of two, two-minute sets of Grade III anterior-to-posterior talocrural joint mobilizations with a one-minute rest between sets. With the participant in a long-sitting position, large-amplitude, one-second rhythmic oscillations from the mid- to end ROM with translation taken to tissue resistance were delivered.16, 18 Plantar massage consisted of two, two-minute sets with a one minute rest between sets. A combination of petrissage and effleurage was delivered to the entire plantar aspect of the foot with the participant supine but no effort was made to constrain the time spent using either technique of the location of the techniques.17, 18 Calf stretching consisted of two stretching sets that were made up of three, 30-second stretches with the knee bent. A ten-second rest was required between stretches while a one-minute rest was required between sets. An adjustable slant board was used set so that the calf was gently stretched.18
Data included participant characteristics, the condition of CAI, as well as dorsiflexion ROM, postural control, and self-assessed function at baseline. Demographics such as age, height, weight, and sex represent participant characteristics, while data related to the condition of CAI included: “yes” responses on the AII, number of lateral ankle sprains sustained, time since the most recent lateral ankle sprain, and number of giving way episodes within the past 6-months. The baseline assessment of the RCT’s primary and secondary outcomes as well as the between limb differences of these outcomes were also entered as independent variables. Patient-oriented outcome measures included self-reported function which was recorded using the FAAM and FAAM-S. For both the FAAM and FAAM-S, a lower percentage represented greater levels of self-reported disability. A participant’s level of self-reported function was recorded for both the treatment and non-treatment limbs. Participants also reported the number of giving way episodes experienced. Giving way was operationally defined as “the regular occurance of uncontrolled and unpredictable episodes of excessive inversion of the rear foot, which do not result in an acute ankle sprain”.25 The weight bearing lunge test (WBLT) assessed dorsiflexion ROM on both limbs.26 The WBLT has been shown to have excellent inter-rater and intra-rater reliability.26 Three, 20-second single limb balance test (SLBT) trials assessed balance on both limbs. Participants stood on a firm surface with their eyes closed. The research team observed each participant for specific balance errors such as opening their eyes, excessive trunk movement, removing their hands from hips, touching down with the contralateral limb, and remaining out of the test position for >5-seconds.27 Previous research has demonstrated good inter-tester reliability for the SLBT.27 The STARS RCT manual of operating procedures, which was approved by the NIAMS and the Data Safety Officer, maintained consistency among test sites.
First, participants were dichotomized as having a successful or unsuccessful treatment at the end of the 2-week intervention based on their SLBT score improvement. A successful treatment response was a change in the SLBT score that exceeded the minimum detectable change (MDC) score calculated from the reliability estimates.28 The MDC score determines the amount of change needed to go beyond the typical measurement error for the outcome. The SLBT change calculated within the treatment limb as described above was then compared to the MDC calculated using the non-treatment limb change from baseline to the 2-week assessment point (1 error).
Individual independent variables were then tested for a univariate relationship with the MDC reference criteria. To do this, we used independent samples t-tests for continuous variables (e.g. height) and X2 tests for categorical variables (e.g. sex) to compare the pre to post change scores of the successful vs. unsuccessful groups.22, 23 Variables with a significance level of p≤0.10 were retained as potential prediction variables.23 A liberal significance level was purposefully chosen to reduce the chance that potential predictor variables would be overlooked. Previous research has used p values as high as 0.20 to identify potential predictor variables.22 A stepwise logistic regression model was then used to determine the most accurate set of potential predictor variables for predicting success of each treatment.22, 23 To again minimize the risk of excluding variables that could help strengthen the model, a significance level of 0.10 was required for removal from the equation. Retained variables were obtained as the CPR.
All potential predictor variables were also submitted to a receiver operator characteristic (ROC) curve analysis.22, 23 Sensitivity, specificity, and positive and negative likelihood ratios (LR) were then calculated for identified cut-off scores for each potential predictor variable. Similarly, diagnostic accuracy and probability of success was calculated for the various combination of predictor variables. Post-test probability was calculated using the +LR and assumed a pre-test probability equal to the number of participants who exceeded the MDC before stratification.
Results
After receiving ankle joint mobilization, 9 of 20 participants (45%) were deemed to have a successful treatment and averaged a 1.77±0.52 reduction in SLBT errors relative to the unsuccessful group (11 of 20 participants [55%], −0.06±0.84 reduction). After receiving plantar massage, 10 of 19 participants (53%) were deemed to have a successful treatment and averaged a 2.63±0.81 reduction in SLBT errors relative to the unsuccessful group (9 of 19 participants [47%], 0.11±0.91 reduction). After stretching, 4 of 20 participants (20%) were deemed to have a successful treatment and averaged a 1.91±0.56 reduction in SLBT errors relative to the unsuccessful group (16 of 20 participants [80%], 0.04±0.89). The univariate comparisons of the potential predictor variables between success and unsuccessful groups sorted by treatment group can be seen in Table 1.
Univariate variables that were retained after the step-wise regression for SLBT success can be seen in Table 2. Combinations of predictor variables, when appropriate, can also be seen in Table 2. As illustrated in Tables 2, there are both individual and combinations of individual predictor variables that demonstrate large and meaningful +LR and high post-test probability percentages for improving SLBT scores following ankle joint mobilizations and plantar massage. While individual variables were identified as potential predictors of calf stretching success, none of these variables were retained in the regression model and thus no combination of variables were able to be calculated.
Table 2.
Sensitivity, specificity, positive likelihood ratio and post-test probability for individual predictor variables and predictor variable combinations, when applicable, sorted by treatment group
| Sensitivity (95% CI) | Specificity (95% CI) | Positive Likelihood Ratio (95% CI) | Posttest Probability | |
|---|---|---|---|---|
| Joint Mobilizations (Pre-Test Probability= 45%) | ||||
|
| ||||
| FAAM between limb difference >11.3%† | 55% (27 to 81%) | 82% (52 to 95%) | 3.05 (0.77 to 12.18) | 71% |
| SLBT errors ≥3† | 33% (12 to 65%) | 91% (62 to 98%) | 3.67 (0.46 to 29.49) | 75% |
| Combination of SLBT errors ≥3 and FAAM between limb difference >11.3% | 100% (21 to 100%) | 58% (36 to 77%) | 2.4 (2.22 to 4.32) | 66% |
|
| ||||
| Plantar Massage (Pre-Test Probability= 53%) | ||||
|
| ||||
| FAAM between limb difference <16.07%† | 100% (72 to 100%0 | 67% (35 to 88%) | 3.0 (1.17 to 7.48) | 77% |
| FAAM-S >70.31% | 90% (60 to 98%) | 33% (12 to 65%) | 1.35 (0.81 to 2.23) | 60% |
| FAAM-S between limb difference <29.69% | 70% (40 to 98%) | 89% (57 to 98%) | 6.30 (0.95 to 41.78) | 88% |
| SLBT ≥3 errors† | 70% (40 to 98%) | 89% (57 to 98%) | 6.30 (0.95 to 41.78) | 88% |
| SLBT between limb difference ≤1 error | 100% (72 to 100%0 | 67% (35 to 88%) | 3.0 (1.17 to 7.48) | 77% |
| Combination of SLBT ≥3 errors and FAAM between limb difference <16.07% | 83% (55 to 95%) | 100% (65 to 100%) | 83.33 (11.66 to 595.48)* | 99%** |
|
| ||||
| Calf Stretching (Pre-Test Probability= 25%) | ||||
|
| ||||
| Height ≥173.35cm | 100% (51 to 100%) | 69% (44 to 86%) | 3.20 (2.22 to 7.06) | 51% |
| Weight ≥79.32kg | 75% (30 to 95%) | 81% (57 to 93%) | 4.00 (1.24 to 12.84) | 57% |
| Total number of sprains ≥11 | 50% (15 to 85%) | 100% (81 to 100%) | 50.00 (5.63 to 443.45)* | 94%** |
| FAAM between limb difference ≤ 25.5% | 100% (51 to 100%) | 69% (44 to 86%) | 3.20 (2.22 to 7.06) | 51% |
+LR was calculated with a specificity of 99% because the actual 100% specificity does not allow for the calculation of the +LR.
Calculated using the modified +LR.
retained in regression model
Discussion
The most important finding is that there were specific predictor variables that substantially improved the prediction of a successful outcome in those participants who received plantar massage. The positive likelihood ratios for both the single and multiple predictors in this group were extremely large. By contrast, while there were predictors of success in the joint mobilization and stretching groups, the predictive ability of these characteristics resulted in very small, but potentially meaningful positive likelihood ratios. These findings indicate that not only does plantar massage appear to be the sensory-targeted strategy for optimizing single limb balance, but there are specific indicators of treatment success.
Because CAI is such a heterogeneous condition and previous research has focused on the changes of group means as a result of intervention,13, 15–17 the development of CPRs for the treatment of CAI would assist clinical decision making and address the growing need to treat a participant’s specific impairments.24 Unfortunately, there is no empirical evidence to help clinicians know which participant characteristics and assessment findings are predictive of successfully treating balance impairments in those with CAI. The primary findings of this secondary analysis indicate that a number of individual and combinations of individual predictor variables can predict when ankle joint mobilizations, plantar massage, or calf-stretching have a high likelihood of successfully improving balance in those with CAI. Most CPR research has based treatment success on a subjective measure of global improvement in self-reported function.22, 23 Using such an outcome, “Do you feel your ankle is more stable since participating in this study?”, pre-test probabilities from the RCT 18 were extremely high for Joint Mobilizations (90%), Massage (84%), and Stretching (75%). Due to the high pre-test probabilities, we defined treatment success on whether or not SLBT score improvements exceeding the calculated MDC for this outcome. Using this approach, our pre-test probabilities among the treatment groups ranged between 20 and 53% and illustrate the disconnect between a participant’s global sense of improvement and actual changes in clinician-oriented outcomes.29
While the results of this investigation are preliminary, they demonstrate that easily acquired demographic and injury information, as well as simple clinician-oriented assessments can aid clinicians in determining if STARS will improve balance in their CAI participants. For example, if a participant with CAI has ≥3 errors during a single limb balance test, they have a 75% chance of having a meaningful improvement in balance following the joint mobilization intervention described. This represents a 30% shift in probability of treatment success just by performing a SLBT assessment before treatment begins. Contrasting this to plantar massage, the same participant would have an 88% improvement, a 35% shift in probability. Combining that the SLBT results with an asymmetry on the FAAM of <16% would result in an almost guarantee (99% probability) that such a participant would benefit from plantar massage. By identifying the sensorimotor impairment (poor single limb balance) and matching it with the appropriate treatments, we can dramatically enhance the likelihood of treatment success.
Calf stretching did not produce substantial predictor combinations that resulted in meaningful shifts in treatment success. However, screening CAI participants based on individual variables would improve treatment success by at least 25% (Table 2). The most robust of these variables was, total number of ankle sprains. Indeed, participants with at least 11 total ankle sprains were highly likely (94%) to have balance improvements following the two-week calf stretching protocol. These stretching results should be interpreted with caution due to the small sample size of participants who had a successful treatment (n=4) and the extreme cut-off score identified for the number of ankle sprains (n=11) needed to predict treatment success. However, these findings provide a much more robust interpretation of the utility of calf-stretching when compared to the group mean findings from the RCT. In the RCT, the calf stretching group demonstrated large effect sizes with confidence intervals that did not cross zero for the changes on the SLBT relative to controls.18 While this is certainly an important finding, it is only through the Bayesian approach of this secondary analysis, that we now know that only 4 of the 20 participants actually demonstrated benefit.
This secondary analysis was conducted as a preliminary investigation into CPRs for improving balance in those with CAI following ankle joint mobilization, plantar massage, and calf stretching. Because this data came from an RCT,18 the sample size was not powered for detecting individual predictor variables as has been identified as a potential limitation previously.22 As a result, post-test probabilities may be artificially inflated. Thus, these results should be interpreted as preliminary until further research has been able to reexamine all variables including those not identified as potential predictor variables22 as confirmation is needed.22, 23 It is also important to note that clinicians have other options to improve balance in CAI patients seeking care if their patients do not meet the criteria reported in this investigation. Perhaps the most effective, based on group mean improvements, is balance training. While clinical prediction rules to help determine which CAI patients are most likely to have meaningful balance improvements following balance training are not yet available, meta-analyses and systematic reviews have demonstrated that balance training improves balance30 and reduces the risk of recurrent injury.13,14 Similarly, combining balance training with STARS may further enhance the individual benefits of both treatments as highlighted in a preliminary investigation.31
As mentioned above, we noted disconnect between a participant’s global sense of improvement and actual changes in clinician-oriented outcomes. However, patient reported outcomes are of the utmost importance8 and we have previously reported meaningful improvements in self-reported function following these same STARS protocols.18 Given their importance, we explored the characteristics that might predict concurrent improvements in both balance and self-reported function (i.e. a FAAM-S change that exceeded the minimally clinically important difference of the FAAM-S (9%) using identical statistical techniques. Pre-Test probability was decreased across all groups (Table 3) as was expected given the more complex criteria for a successful treatment which is why this was an exploratory analysis and not the primary aim of this investigation. These results should be interpreted with caution given the low number of participants with a successful treatment in each group (Mobilizations: 7 of 20 participants; Massage: 5 of 20 participants; Stretching: 3 of 20 participants). However, the individual predictor variables and combinations of those individual predictor variables demonstrated large shifts in the probability of a successful treatment after stratification (Mobilizations: Δ=61%; Massage: Δ=71%; Stretching: Δ=77%, see Table 3). These shifts were calculated as the difference between the post and pre-test probability values for each treatment. This exploratory analysis supports the initial findings as seen in Table 2 as similar predictor variables were identified. The large shifts in the probability of successfully improving both self-reported function and balance after stratification provide evidence of the contextual dependence of patient- and clinician-oriented outcomes. However, the preliminary nature of this sample highlights the need for large scale investigations aimed at developing CPRs for those with CAI.
Table 3.
Factors depicting the contextual dependence of patient- and clinician-oriented variables. Sensitivity, specificity, positive likelihood ratio and post-test probability for individual predictor variables and predictor variable combinations, sorted by treatment group with success defined as improvements in self-reported function and improvements in balance.
| Sensitivity (95% CI) | Specificity (95% CI) | Positive Likelihood Ratio (95% CI) | Posttest Probability | |
|---|---|---|---|---|
| Joint Mobilizations (Pre-Test Probability= 35%) | ||||
| SLBT errors ≥3† | 86% (49 to 97%) | 69% (42 to 87%) | 2.79 (1.2 to 6.98) | 60% |
| FAAM between limb difference >14.9%† | 57% (25 to 84%) | 92% (67 to 99%) | 7.43 (1.37 to 44.49) | 80% |
| Combination of SLBT errors ≥3 and FAAM between limb difference >14.9% | 43% (16 to 75%) | 100% (77 to 100%) | 42.86 (6.48 to 271.36)* | 96%** |
|
| ||||
| Plantar Massage (Pre-Test Probability= 26%) | ||||
|
| ||||
| AII “yes” responses ≤6† | 100% (57 to 100%) | 64% (39 to 84%) | 2.8 (1.9 to 6.1) | 49% |
| SLBT between limb difference ≤2 errors† | 80% (38 to 96%) | 100% (79 to 100%) | 80 (13.34 to 466.24)* | 97%** |
| Combination of AII “yes” responses ≤6 and SLBT between limb difference ≤2 errors | 80% (38 to 96%) | 100% (79 to 100%) | 80 (13.34 to 466.24)* | 97%** |
|
| ||||
| Calf Stretching (Pre-Test Probability= 15%) | ||||
|
| ||||
| Total number of sprains ≥11† | 67% (21 to 94%) | 100% (82 to 100%) | 66.67(9.47 to 411.69)* | 92%** |
+LR was calculated with a specificity of 99% because the actual 100% specificity does not allow for the calculation of the +LR.
Calculated using the modified +LR.
retained in regression model
Conclusion
This preliminary investigation demonstrates a number of individual and combinations of variables can predict whether or not a participant with CAI could have a meaningful improvement in a clinician-oriented balance outcome following 6, 5-minute sessions of ankle joint mobilization, plantar massage, and calf stretching over a 2-week period. The post-test probability score improvements following the stratification of participants based on these predictor variables was a minimum of 24% and as large as 69% in one situation. This investigation serves as a basis for further studies and may be used by clinicians as a guide to assist with clinical decision in treating those with CAI using an Impairment-Based rehabilitation model.
Practical Implications.
CAI participants who make ≥3 errors during a SLBT have a 75% chance of having a meaningful improvement in balance following 6, 5-minute ankle joint mobilization sessions over a 2-week period of time.
CAI participants who make ≥3 SBLT errors and have an asymmetry on the FAAM of <16% have 99% probability of a meaningful balance improvement following 6, 5-minute plantar massage sessions over a 2-week period.
CAI participants with at least 11 total ankle sprains have a 94% probability of having meaningful balance improvements following 6, 5-minute calf stretching sessions over a 2-week period.
Acknowledgments
This study was funded by the NIAMS (1R03AR061561).
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Fong DT, Hong Y, Chan LK, et al. A systematic review on ankle injury and ankle sprain in sports. Sports Med. 2007;37(1):73–94. doi: 10.2165/00007256-200737010-00006. [DOI] [PubMed] [Google Scholar]
- 2.Waterman BR, Owens BD, Davey S, et al. The epidemiology of ankle spranis in the united states. J Bone Joint Surg Am. 2010;92(13):2279–2284. doi: 10.2106/JBJS.I.01537. [DOI] [PubMed] [Google Scholar]
- 3.McKay GD. Ankle injuries in basketball: injury rate and risk factors. Br J Sports Med. 2001;35:103–108. doi: 10.1136/bjsm.35.2.103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Anandacoomarasamy A, Barnsley L. Long term outcomes of inversion ankle injuries. Br J Sports Med. 2005;39(3):1–4. doi: 10.1136/bjsm.2004.011676. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Attenborough AS, Hiller CE, Smith RM, et al. Chronic ankle instability in sporting populations. Sports Med. 2014;44(11):1545–1556. doi: 10.1007/s40279-014-0218-2. [DOI] [PubMed] [Google Scholar]
- 6.Mandarakas M, Pourkazemi F, Sman A, et al. Systematic review of chronic ankle instability in children. J Foot Ankle Res. 2014;7(1):21. doi: 10.1186/1757-1146-7-21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Gribble PA, Delahunt E, Bleakley CM, et al. Selection criteria for patients with chronic ankle instability in controlled research: a position statement of the international ankle consortium. J Orthop Sports Phsy Ther. 2013;43(8):585–591. doi: 10.2519/jospt.2013.0303. [DOI] [PubMed] [Google Scholar]
- 8.Wikstrom EA, Hubbard-Turner T, McKeon PO. Understanding and treating lateral ankle sprains and their consequenes: a constraints-based approach. Sports Med. 2013;43(6):385–393. doi: 10.1007/s40279-013-0043-z. [DOI] [PubMed] [Google Scholar]
- 9.Houston MN, Hoch JM, Hoch MC. Patient-reported outcome measures in individuals with chronic ankle instability: a systematic review. J Athl Train. 2015;50(10):1019–1033. doi: 10.4085/1062-6050-50.9.01. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hubbard-Turner T, Turner MJ. Physical activity levels in college students with chronic ankle instability. J Athl Train. 2015;50(7):742–747. doi: 10.4085/1062-6050-50.3.05. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hintermann B, Boss A, Schafer D. Arthroscopic findings in patients with chronic ankle instability. Am J Sports Med. 2002;30(3):402–409. doi: 10.1177/03635465020300031601. [DOI] [PubMed] [Google Scholar]
- 12.Wang HK, Chen CH, Shiang TY, et al. Risk-factor analysis of high school basketball-player ankle injuries: a prospective controlled cohort study evaluating postural sway, ankle strength, and flexibility. Arch Phys Med Rehabil. 2006;87(6):821–825. doi: 10.1016/j.apmr.2006.02.024. [DOI] [PubMed] [Google Scholar]
- 13.Schiftan GS, Ross LA, Hahne AJ. The effectiveness of proprioceptive training in preventing ankle sprains in sporting populations: A systematic review and meta-analysis. J Sci Med Sport. 2015;18(3):238–244. doi: 10.1016/j.jsams.2014.04.005. [DOI] [PubMed] [Google Scholar]
- 14.McGuine TA, Keene JS. The effect of a balance training program on the risk of ankle sprains in high school athletes. Am J Sports Med. 2006;34(7):1103–1111. doi: 10.1177/0363546505284191. [DOI] [PubMed] [Google Scholar]
- 15.Ross SE, Arnold BL, Blackburn JT, et al. Enhanced balance associated with coordination training with stochastic resonance stimulation in subjects with functional ankle instability: an experimental trial. J NeuroEngineering and Rehab. 2007;4(47) doi: 10.1186/1743-0003-4-47. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Hoch MC, Mullineaux DR, Andreatta RD, et al. Effect of a 2-week joint mobilization intervention on single-limb balance and ankle arthrokinematics in those with chronic ankle instability. J Sport Rehab. 2014;23(1):18–26. doi: 10.1123/jsr.2012-0125. [DOI] [PubMed] [Google Scholar]
- 17.LeClaire JE, Wikstrom EA. Massage for postural control in those with chronic ankle instability. Athl Train Sports Health Care. 2012;4(5):213–219. [Google Scholar]
- 18.McKeon PO, Wikstrom EA. Sensory-targeted ankle rehabilitation strategies for chronic ankle instability. Med Sci Sports Exer. 2015 doi: 10.1249/MSS.0000000000000859. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.McGinn TG, Guyatt GH, Wyer PC, et al. Users’ guides to the meduical literatuer: XXII: how to use articles about clinical decision rules. JAMA. 2000;284:79–84. doi: 10.1001/jama.284.1.79. [DOI] [PubMed] [Google Scholar]
- 20.Laupacis A, Sekar N, Stiell IG. Clinical prediction rules. a review and suggested modification of methodological standards. JAMA. 1997;277:488–494. [PubMed] [Google Scholar]
- 21.Brennan GP, Fritz JM, Hunter SJ, et al. Identifying subgroups of patients with acute/subacute “nonspecific” low back pain: results of a randomized clinical trial. Spine. 2006;31:623–631. doi: 10.1097/01.brs.0000202807.72292.a8. [DOI] [PubMed] [Google Scholar]
- 22.Vicenzino B, Collins N, Cleland J, et al. A clinical prediction rule for identifying patients with patellofemoral pain who are likely to benefit from foot orthoses: a preliminary determination. Br J Sports Med. 2010;44:862–866. doi: 10.1136/bjsm.2008.052613. [DOI] [PubMed] [Google Scholar]
- 23.Whitman JM, Cleland JA, Mintken P, et al. Predicting short-term response to thrust and nonthrust manipulation and exercise in patients post inversion ankle sprain. J Orthop Sports Phys Ther. 2009;39(3):188–200. doi: 10.2519/jospt.2009.2940. [DOI] [PubMed] [Google Scholar]
- 24.Donovan L, Hertel J. A new paradigm for rehabilitation of patients with chronic ankle instability. Phys Sportsmed. 2012;40(4):41–51. doi: 10.3810/psm.2012.11.1987. [DOI] [PubMed] [Google Scholar]
- 25.Gribble PA, Delahunt E, Bleakley CM, et al. Selection criteria for patients with chronic ankle instability in controlled research: a position statement of the international ankle consortium. J Orthop Sports Phys Ther. 2013;43(8):585–591. doi: 10.2519/jospt.2013.0303. [DOI] [PubMed] [Google Scholar]
- 26.Bennell KL, Talbot RC, Wajswelner H, et al. Intra-rater and inter-rater reliability of a weightbearing lunge measure of ankle dorsiflexion. Aust J Physiother. 1998;44(3):175–180. doi: 10.1016/s0004-9514(14)60377-9. [DOI] [PubMed] [Google Scholar]
- 27.Riemann BL, Guskiewicz KM, Shields WW. Relationship between clinical and forceplate measures of postural stability. J Sport Rehab. 1999;8:71–82. [Google Scholar]
- 28.McKeon PO, Wikstrom EA. The responsiveness of multifactorial outcome measures for those with chronic ankle instability. Br J Sport Med. 2015;49(S1):A20. [Google Scholar]
- 29.McKeon PO. First Words: Cultivating functional variability: The dynamical systems approach to rehabilitation. Athl Ther Today. 2009;14(4):1–3. [Google Scholar]
- 30.Wikstrom EA, Naik S, Lodha N, Cauraugh JH. Balance capabilities after lateral ankle trauma and intervention: a meta-analysis. Med Sci Sports Exerc. 2009;41(6):1287–1295. doi: 10.1249/MSS.0b013e318196cbc6. [DOI] [PubMed] [Google Scholar]
- 31.Burcal CJ, Trier AY, Wikstrom EA. Balance training vs. balance training with STARS in CAI patients: a raondomized controlled trial. J Sport Rehab. 2016 doi: 10.1123/jsr.2016-0018. http://dx.doi.org/10.1123/jsr.2016-0018. [DOI] [PubMed]