Prediction of Recurrent Lateral Ankle Sprain and Ankle Pain Using Applied Care Strategies and Patient-Reported Outcomes

We used a prospective cohort design in which we enrolled 63 individuals (women = 38, men = 25, 22.3 ± 9.6 years, 172.1 ± 12.2 cm, 75.3 ± 17.1 kg) who sustained an acute LAS within 1 week before beginning the study. All participants were recruited from a large university and the surrounding community. Participants responded to approved advertising from the university’s clinical and translational research center, approved flyers in university and local medical facilities and campus and community recreational centers, or by word of mouth.

All procedures were completed electronically through email and online survey links. The study was approved by the University of Kentucky institutional review board, and participants provided consent electronically. Following enrollment, participants were emailed a survey link regarding health history and severity of the LAS that they sustained. More specifically, diagnosis of LAS was classified as a traumatic injury to the lateral ankle ligaments following a combination of ankle inversion, plantarflexion, and/or internal rotation motion that was diagnosed by a health care provider and resulted in at least 1 day of activity restriction. Furthermore, we asked participants to recall how their health care provider classified their LAS severity, which is typically based on a 3-tiered categorization (ie, mild, moderate, and severe; grade 1, grade 2, and grade 3). Participants then received a weekly survey link asking them to self-report their LAS recovery, specifically asking if they had returned to their preinjury level of physical activity capacity. Once the participants indicated that they had reached RTA, they completed online surveys accessed through email links they received at the following times: RTA, 6 months post-RTA, and 12 months post-RTA. Study data were collected and managed using REDCap electronic data capture tools hosted at the University of Kentucky. At RTA, participants answered questions regarding demographics (age, sex, height, and mass) and whether they used applied care strategies (crutches or a walking boot and supervised or unsupervised rehabilitation). Supervised rehabilitation was classified as therapy provided directly by a rehabilitation specialist (ie, athletic trainer or physical therapist) within a rehabilitation clinic. Unsupervised rehabilitation was classified as home-based care that was not directly overseen by a rehabilitation specialist. Participants also completed the following PRO surveys at RTA:

1. The Foot and Ankle Disability Index (FADI) is a 34-item measure of ankle-specific self-reported function and pain. It includes 3 separated subscales related to activities of daily living (FADI-ADL), pain (FADI-P), and sport (FADI-S). Each subscale is measured from 0 to 100%, with higher scores equating to better function or lower pain intensity. The FADI subscales have previously demonstrated strong test-retest reliability (interclass correlation coefficient [ICC] = 0.84–0.93).¹⁴ 

2. The Godin Leisure-Time Exercise Questionnaire quantifies an individual’s typical frequency of participation in strenuous, moderate, and mild exercise over a 7-day period. Higher scores equate to a greater physical activity level. This questionnaire previously demonstrated moderate to excellent reliability (ICC = 0.46–0.94).¹⁵ 

3. The Short Form 8 (SF8) is an abbreviated version of the Short Form-36 and is used to assess HRQoL. Scores on the SF8 surveys were separated into Body Pain and General, Physical, and Mental components. The SF8 previously demonstrated moderate to good reliability (ICC = 0.59–0.74).¹⁶ 

At the 6-month interval, participants completed the same series of PRO surveys. At the 12-month interval, participants were asked if they sustained any additional LAS since RTA and, if yes, how many they had in the 12-month period and when was the most recent LAS. The presence of ankle pain at each interval was a dichotomous outcome determined by the FADI-P (pain versus no pain).

We used Fisher exact tests to determine if recurrent LAS and presence of ankle pain at a 12-month follow-up were related to use of protective devices, use of supervised rehabilitation, or the presence of ankle pain at RTA. Independent t tests compared participant demographics and PRO scores at RTA and 6-month follow-up between participants who did and did not sustain a recurrent LAS or experience ankle pain at the 12-month follow-up. We assessed the magnitude of significant differences with Cohen d effect sizes (negligible < 0.20, small = 0.20–0.49, moderate = 0.5–0.79, large ≥ 0.8).

Backward logistic regression analyses determined if sustaining a recurrent LAS and experiencing ankle pain at 12 months could be predicted by participant demographics, use of protective devices, use of physical rehabilitation, ankle pain at RTA, and PRO scores at RTA and 6-month follow-up. In the event of a significant logistic regression model, we used receiver operating characteristic (ROC) curve analysis to measure the area under the ROC curve and further analyze the predictive value of the model. The level of significance was set a priori at P < .05. All statistics were conducted using IBM SPSS (version 27; IBM Corp).

We received incomplete responses for some participants, leading to reduced samples when analyzing data related to physical rehabilitation, recurrent LAS, and ankle pain (Table 1). We compared primary demographics between participants who did and did not complete 12-month surveys using independent t tests but found no significant differences (Table 2). Fisher exact tests revealed that participants who used a walking boot less frequently sustained a recurrent LAS within 12 months (Table 3). There were no other significant differences in proportions of applied care strategies among participants who did and did not sustain a recurrent LAS or experience persistent ankle pain. Independent t tests indicate that participants who did and did not sustain a recurrent LAS differed in age and height (Table 4). There were no significant differences between participants who experienced ankle pain at 12 months (Table 5).

Logistic regression analysis determined that the combination of increased height and lack of walking boot use were significant predictors of recurrent LAS (χ²₍₂₎ = 11.14, P < .01, R² = 0.33). These variables exhibited predictive value in the ROC curve analysis (area under the ROC curve = 0.81 [0.68, 0.95], P < .01; Figure). There were no significant predictors of ankle pain at 12 months.

The purpose of this study was to determine if recurrent LAS and ankle pain could be predicted by applied care strategies and PRO scores in patients with an acute LAS. Our primary finding was that recurrent LAS was predicted by combinations of factors. Specifically, the use of a walking boot before RTA contributed to reduced odds of sustaining a recurrent LAS within 12 months of the acute LAS. Thus, a brief immobilization period via a walking boot provided protection against a common consequence of acute LAS. Historically, whether a period of immobilization or early mobilization is more beneficial after LAS has been an area of contention. Earlier studies reported that functional bracing and mobilization resulted in quicker RTA and lower short-term symptoms than multiple weeks of complete immobilization.^10,11  Later, others reported that immobilization via external supports resulted in better self-reported function, pain, and symptoms than no immobilization, particularly when applied to severe ankle sprains.¹⁷  A more recent animal model demonstrated that using 1 to 2 weeks of rest as a form of immobilization following severe LAS led to more favorable balance performance during weeks 42 to 54 after injury.¹²  Collectively, the current and previous findings indicate early mobilization results in more favorable outcomes for the first few weeks or months after LAS, but brief immobilization might result in better long-term outcomes, especially among patients with a severe LAS. Although we asked about our participants’ injury severity, we received only 38 responses to this question. Of those, 14 did not know their severity, whereas 8 had a grade 1 LAS, 13 had a grade 2 LAS, and 3 had a grade 3 LAS as indicated to the patient by their health care provider. Unfortunately, these incomplete data prevent us from exploring how severity could interact with the potential benefit of early immobilization, but this is an important factor that should be explored in future work.

In addition to a lack of walking boot use, we found that taller participants were also more likely to sustain a recurrent injury within 12 months of a LAS. A large effect size provides further evidence of a height disparity between participants with and without a recurrent LAS. Greater height was previously implicated as a risk factor for acute and recurrent ankle sprains among military, athletic, and general populations.^6,7,18  Others explained that longer (and heavier) body segments of taller individuals possess greater inertia and therefore may demand greater force production to change momentum and counteract injurious moments at the ankle.^6,18  The average height of our participants with a recurrent LAS (180.3 ± 10.6 cm) was comparable to military recruits (179.5 ± 6.5 cm) with an acute LAS, indicating that individuals of similar or greater height might have elevated risk for primary and recurrent LAS.¹⁸  However, others reported higher (191.0 cm) and lower (173.0 cm) height thresholds as risk factors for recurrent LAS in athletic and general populations, respectively.^6,7  Although height is not a modifiable factor, this is still an important finding. Clinicians may need to be cognizant of monitoring the recovery of their taller patients, and future work needs to determine if more comprehensive follow-ups are needed for this taller population of LAS patients.

Although we found effective predictors of recurrent LAS, none of our outcomes had predictive utility for ankle pain 12 months after an acute LAS. While ankle pain is not currently used as a defining characteristic of CAI, recent studies have highlighted the relevance of pain to the CAI population.¹⁹  A systematic review reported that ankle pain is present in up to 79% of individuals with CAI.²⁰  Further, over two-thirds of individuals with CAI experience pain for several months or years after LAS.²⁰  Although most of those with CAI experience mild to moderate pain levels, the existence of pain is associated with higher levels of perceived ankle instability and lower levels of self-reported function.^20–22  Joint pain can also contribute to altered afferent neural signaling, which ultimately leads to central nervous system alterations and the presence of arthrogenic muscle inhibition.²³  With awareness of the growing prevalence of ankle pain among ankle-injured populations, more work is needed to determine which patients will experience long-term ankle pain following LAS. Others have discovered that psychological constructs such as anxiety and posttraumatic stress are predictors of chronic pain after musculoskeletal injuries.²⁴  Although we did not find the SF8 Mental Component score to be a predictor of pain or recurrent LAS, future studies might consider analyzing other psychological impairments that follow LAS, such as injury-related fear and pain catastrophizing, as potential predictors of ankle pain.

We found that none of the PROs collected at RTA or 6 months were predictive of recurrent LAS or ankle pain at the 12-month follow-up. This is somewhat surprising because others previously identified ankle-specific PROs as predictors of recurrent LAS.⁶  However, those authors used the Cumberland Ankle Instability Tool as a predictor, whereas we did not. Although reduced self-reported function measured with the FADI has exhibited predictive utility for CAI, others reported that neither self-reported function nor pain at RTA were predictive of recurrent LAS.^2,7  Although these previous findings agree with ours, we must note that participants in the previous study had levels of self-reported function consistent with CAI at RTA, whereas those in the current study had FADI scores above thresholds for CAI, indicating that our sample was not experiencing self-reported dysfunction.¹⁹  Therefore, our participants might have possessed little impairment that would predispose them to recurrent LAS or persistent ankle pain.

To our knowledge, no previous authors have attempted to predict recurrent LAS or ankle pain using measures of HRQoL or physical activity. An earlier study reported differences in general HRQoL between individuals with and without LAS history via the SF36, and others found that individuals with CAI scored worse on the SF36 Physical Component score than healthy control individuals.^3,25  A more recent study observed worse scores on the SF8 Physical Component score in adults with self-reported ankle sprain history.²⁶  Several other studies have reported that individuals with a previous LAS or CAI are less physically active than healthy individuals.^27,28  However, research in this area suggests that acute ankle sprains cause a progressive physical activity decline that emerges in years following the acute injury.^27,28  Therefore, perhaps physical activity within 6 months of RTA is unable to predict recurrent LAS or ankle pain. There are many methods of assessing physical activity beyond the selected measures in this study, and continued investigation may be needed to determine the impact of LAS on short-term physical activity and subsequently any potential association with developing recurrent LAS issues.

According to our findings, clinicians treating patients with an acute LAS should provide a walking boot to immobilize the injured joint. However, our study design does not allow us to recommend an exact duration for walking boot use; but generally, a brief period of immobilization appears to protect against recurrent LAS. Among those who seek medical care for a LAS, only about 8% are prescribed a walking boot, and less than 10% receive any form of immobilization or assistive ambulation device.²⁹  This lack of immobilization provided to patients might contribute to long-term consequences of LAS. In addition to walking boot prescription, clinicians should consider the height of LAS patients as a risk factor for recurrent LAS. Although height is clearly limited as a nonmodifiable risk factor, taller patients might benefit from more conservative treatment that can mitigate risk of recurrent LAS.

This study had several limitations. First, several participants were lost to follow-up, which reduced the data available at the 12-month interval that could have been available. Our use of surveys to collect data caused us to rely on participants’ recall of injury characteristics and applied care strategies. Using self-reported data could introduce risk for recall bias due to potential inconsistencies in individuals’ abilities to remember details of their medical history. To avoid recall bias, future studies should obtain injury and care data directly from electronic medical records instead of participants themselves. Another potential solution is to include athletic trainers or other clinicians directly in the study procedures by having them document relevant data regarding the injury and treatment. Our study was further limited by our use of outcomes that contained only patient perceptions. Future studies should aim to include clinician-oriented outcomes, which have previously demonstrated ability to predict recurrent LAS and CAI.^2,6  We also did not ask probing questions regarding the use of care strategies. The duration of immobilization device use and specific parameters of rehabilitation protocols could have contributed to stronger prediction models.

Patients who did not undergo a brief period of immobilization with a walking boot had greater odds of sustaining a recurrent LAS within 12 months of an acute LAS. Additionally, taller patients had greater risk of recurrent LAS. Clinicians treating patients with LAS should consider using a walking boot to prevent long-term consequences and consider taller patients as having greater risk of recurrent LAS.