Abstract
Background:
The Ankle-GO score is an objective battery for return-to-sports (RTS) decision-making after nonoperative or surgical treatment in patients with chronic ankle instability (CAI). However, direct comparisons of Ankle-GO subscales between CAI patients with and without successful RTS are limited, and predictors of successful RTS remain unclear.
Purpose:
To directly compare the Ankle-GO subscales between CAI patients with and without successful RTS after nonoperative treatment and modified Broström operation (MBO) and to explore which Ankle-GO subscales are significantly associated with successful RTS.
Study Design:
Case control study; Level of evidence, 3.
Methods:
A total of 100 patients (100 ankles) with CAI were enrolled. Of these, 52 ankles underwent nonoperative treatment (31 with successful RTS and 21 without) and 48 ankles underwent MBO (26 with successful RTS and 22 without). Successful RTS was defined as return to preinjury sport level with a Tegner activity score ≥6. The Ankle-GO, which consists of 6 subscales, was compared between patients with and without successful RTS: the single-leg stance test, modified star excursion balance test, side hop test (SHT), figure-of-8 test (F8T), Foot and Ankle Ability Measure, and Ankle Ligament Reconstruction–Return to Sport after Injury (ALR-RSI). Statistical analyses were performed using independent t tests and multivariable binary logistic regression analyses with receiver operating characteristic curve analysis.
Results:
In the nonoperative treatment group, CAI patients with successful RTS demonstrated superior Ankle-GO scores, SHT, and F8T compared with those without successful RTS (all P < .001). In the MBO group, CAI patients with successful RTS demonstrated superior Ankle-GO scores, SHT, F8T, and ALR-RSI scores compared with those without successful RTS (all P < .001). Furthermore, SHT was a significant predictor of RTS in both the nonoperative treatment group (odds ratio [OR], 18.03; 95% CI, 2.8-36.8) and MBO group (OR, 4.59; 95% CI, 2.0-10.7).
Conclusion:
Our study demonstrated that in both the nonoperative treatment and MBO groups, SHT and F8T were key factors distinguishing patients with CAI who had successful RTS from those who did not. In addition, SHT was the strongest independent predictor of RTS in both the nonoperative treatment and MBO groups. Clinicians and therapists should consider these findings when planning rehabilitation and RTS strategies.
Repeated ankle sprains cause chronic ankle instability (CAI), which is characterized by persistent ankle pain, giving way, and dysfunction.5,7 CAI is classified into mechanical ankle instability and functional ankle instability,5,10 which is accompanied by deficits in the neuromuscular system, including muscle performance, proprioception, and altered postural control.10,11,25,26 For this reason, nonoperative treatments such as muscle strengthening and proprioceptive training are considered a first-line approach in the initial management of CAI. However, surgical treatment is often considered when persistent pain and instability remain despite nonoperative management. 16
Nonoperative treatment in patients with CAI—such as muscle strengthening, balance training, and neuromuscular training—has been reported to yield positive clinical and functional outcomes. 28 In addition, surgical treatment for CAI has advanced significantly, and the modified Broström operation (MBO), which involves repair of the anterior talofibular ligament and/or calcaneofibular ligament, is currently considered the standard surgical procedure. 6 Several studies have reported that similar to nonoperative treatment, MBO has favorable clinical and functional outcomes, including improvements in pain, stability, and overall function.11,16 However, despite satisfactory outcomes with both nonoperative and surgical treatments, return-to-sports (RTS) rates have varied widely, ranging from 50%1,13,27 to 90%.13,27 Therefore, several studies have assessed various functional performance abilities, such as muscle strength, static and dynamic balance, and hopping, to identify successful RTS.9,10,25,29 Based on these components, the Ankle-GO score4,19,20 was developed as a representative assessment tool for evaluating RTS after nonoperative and surgical treatment. The Ankle-GO score consists of 4 functional performance tests and 2 patient-reported questionnaires, with a total score of 25 points, 21 and demonstrates strong reliability and validity for assessing successful RTS.4,19,20 However, to our knowledge, few studies have directly compared Ankle-GO scores or subscales between patients who had successful RTS and those who did not. In addition, no study has identified specific subscales that are predictive of successful RTS after nonoperative treatment or MBO. Identifying subscales closely related to RTS success could streamline clinical evaluation, allowing for more focused rehabilitation and efficient assessments.
The primary aim of this study was to directly compare the Ankle-GO scores and their subscales between patients with and without successful RTS within groups with nonoperative treatment or MBO. The secondary aim was to explore whether any Ankle-GO subscales are predictors of successful RTS. The hypothesis of this study was that patients who achieved successful RTS would perform better across all subscales of the Ankle-GO than those who did not.
Methods
Study Design and Participants
This study retrospectively analyzed prospectively collected data from 100 patients (100 ankles) with CAI who received nonoperative treatment or MBO between 2017 and 2025, and it was approved by our institutional review board (No. ED17143).
Inclusion criteria were as follows: (1) patients who had an ankle sprain at least 12 months ago and had had an ankle sprain at least once a year, (2) CAI patients who had mechanical ankle instability or functional ankle instability, (3) patients with CAI with a preinjury Tegner activity score of ≥6, and (4) patients who had Ankle-GO assessment after nonoperative treatment or MBO. CAI was identified from medical records and plain stress radiographs. Ultrasound or magnetic resonance imaging scans were evaluated as needed. MBO was performed when pain or ankle instability persisted despite 2 to 3 months of nonoperative treatment. The exclusion criteria were as follows: (1) patients with a preinjury Tegner activity score <6, (2) CAI patients with ankle fractures, and (3) patients without Ankle-GO assessment data. RTS was evaluated 3 months after the lateral ankle sprain in the nonoperative treatment group and 4 months after surgery in the MBO group, 12 and Ankle-GO assessments were conducted at the same time. In addition, RTS was defined using the Tegner activity scale.2,8,15 A Tegner activity score of ≥6 was defined as performing recreational sports or higher level activities,2,30 which involve participation in sports that place significant stress on the knee or ankle, such as jogging, tennis, soccer, and basketball. 2 Therefore, successful RTS in this study was defined as returning to preinjury levels of sports activity or higher, as indicated by a Tegner activity score of ≥6. Ultimately, among the 100 patients who participated in this study, 52 were classified into the nonoperative treatment group and 48 were classified into the MBO group. The nonoperative treatment group included 31 patients who achieved successful RTS and 21 who did not, whereas the MBO group included 26 patients who achieved successful RTS and 22 who did not.
Outcome Measures
The Ankle-GO consists of 6 subscales: 4 functional performance tests and 2 patient-reported questionnaires.4,21 The functional performance tests include the single-leg stance test (SLS), modified star excursion balance test (mSEBT), side hop test (SHT), and figure-of-8 test (F8T). Patient-reported questionnaires consisted of the Foot and Ankle Ability Measure (FAAM) and Ankle Ligament Reconstruction–Return to Sport after Injury (ALR-RSI). The total score ranges from 0 to 25, with higher scores indicating better ankle function and a lower risk of reinjury.
Functional Performance Test
The SLS measures static postural stability with the participant on a stable surface with eyes closed, standing on 1 leg for 2 trials of 20 seconds each. A maximum of 3 points were awarded for the test.4,21 The mSEBT assesses dynamic postural stability in 3 directions, including the anterior, posteromedial, and posterolateral directions. Measurements from the 3 trials were normalized to the leg length of each patient. A maximum of 7 points were awarded for the test.4,21 The SHT assesses hopping laterally and medially 10 times between 2 lines placed 30 cm apart. A maximum of 5 points were awarded for the test.4,21 The F8T measures the ability to skip on 1 foot over cones placed 5 m apart, in a figure-of-8 pattern for a total distance of 20 m. A maximum of 3 points were awarded for the test.4,21
Patient-Reported Questionnaires
The FAAM questionnaire evaluates physical function in 21 daily activity items and 8 sports-specific items, with higher scores reflecting better functional performance. A maximum of 4 points were awarded for the test.4,21 The ALR-RSI assesses psychological readiness for RTS through 12 items, with higher scores indicating greater mental preparedness. A maximum of 3 points were awarded for the test.4,21
Statistical Analysis
The Shapiro-Wilk test was used to assess whether continuous variable followed a normal distribution. All continuous variables were normally distributed. For comparative analyses between groups, an independent t test was used for continuous variables that followed a normal distribution, and the Fisher exact test was used for categorical variables. To investigate the primary aim, an independent t test was used to compare the means of independent variables between patients with and without successful RTS. To investigate the second aim, a multivariable binary logistic regression analysis was performed to identify independent predictors associated with RTS. All clinically relevant variables, including sex, age, body mass index, Ankle-GO scores, and Ankle-GO subscales (adjust according to variable list), were entered simultaneously into the model. A correlation analysis between variables was conducted to check multicollinearity with the independent variables before regression analysis. The predictive ability of the model was evaluated using receiver operating characteristic curve analysis, with results presented as area under the curve (AUC), sensitivity, and specificity. Data analysis was performed using SPSS (Version 21.0), with the P value set at .05.
Results
Table 1 summarizes the demographic data for the 2 groups (nonoperative treatment, successful RTS vs non-RTS; MBO, successful RTS vs non-RTS). Sex, age, height, weight, body mass index, injured ankle, preinjury Tegner activity levels, and visual analog scale scores did not differ significantly between groups (P > .05), but significant differences were seen in current Tegner activity levels (all P < .001).
Participant Characteristics a
The values are expressed as mean ± SD unless otherwise noted.
Comparison of Outcomes Between the Groups
In the nonoperative treatment group (Table 2), significant differences were found in the overall Ankle-GO score (95% CI, 3.9 to 5.4; P < .001), SHT (95% CI, −3.5 to −2.4; P < .001), and F8T (95% CI, −3.9 to −2.6; P < .001) between the 2 groups, but not in SLS, mSEBT, FAAM, or ALR-RSI (all P > .05), indicating that SHT, F8T, and Ankle-GO overall score were better in patients with CAI who achieved RTS than those who did not.
Ankle-GO Scores Between the Groups a
ADL, activities of daily living; ALR-RSI, Ankle Ligament Reconstruction–Return to Sport after Injury; FAAM, Foot and Ankle Ability Measure; F8T, figure-of-8 test; mSEBT, modified star excursion balance test; SHT, side hop test; SLS, single-leg stance test.
Expressed as mean ± SD.
In the MBO group (Table 2), the overall Ankle-GO score (95% CI, 4.2 to 6.3; P < .001), SHT (95% CI, −5.1 to −3.7; P < .001), F8T (95% CI, −4.6 to −3.4; P < .001), and ALR-RSI (95% CI, 22.6 to 29.3; P < .001) differed significantly between the 2 groups, but not SLS, mSEBT, or FAAM (all P > .05), indicating that SHT, F8T, ALR-RSI, and Ankle-GO overall score were better in patients with MBO who achieved RTS than those who did not.
Logistic Regression Analysis for Predictors of Return to Sports
Table 3 presents the results of logistic regression analysis identifying predictors of RTS.
Logistic Regression Analysis of Ankle-GO Subscales Predicting Successful Return to Sports a
ADL, activities of daily living; ALR-RSI, Ankle Ligament Reconstruction–Return to Sport after Injury; FAAM, Foot and Ankle Ability Measure; F8T, figure-of-8 test; mSEBT, modified star excursion balance test; OR, odds ratio; SHT, side hop test; SLS, single-leg stance test.
In the nonoperative treatment group, Ankle-GO (B = 5.60; Wald = 6.22; P = .013; odds ratio [OR] = 2.49; 95% CI, 1.8 to 24.4) and SHT (B = 2.89; Wald = 9.21; P = .002; OR = 18.03; 95% CI, 2.8 to 36.8) were found to be statistically significant predictors of RTS, but not F8T, SLS, mSEBT, FAAM, or ALR-RSI (all P > .05), indicating that in particular, a higher SHT was strongly associated with an increased likelihood of RTS (AUC = 0.978) (Figure 1A).
Receiver operating characteristic (ROC) curve for side hop test in predicting return to sports. (A) Nonoperative treatment group. (B) Modified Broström operation group.
In the MBO group, Ankle-GO (B = 1.58; Wald = 1.16; P = .010; OR = 1.85; 95% CI, 1.3 to 5.8) and SHT (B = 1.53; Wald = 12.52; P < .001; OR = 4.59; 95% CI, 2.0 to 10.7) were found to be statistically significant predictors of RTS, but not F8T, SLS, mSEBT, FAAM, or ALR-RSI (all P > .05), indicating that in particular, a higher SHT was strongly associated with an increased likelihood of RTS (AUC = 0.995) (Figure 1B).
Discussion
The main finding of the present study was that in the nonoperative treatment and MBO groups, the scores for Ankle-GO, SHT, and F8T were better in patients with CAI who achieved successful RTS than in those who did not. In addition, ALR-RSI was better in patients with MBO who had successful RTS than in those who did not. Both the total Ankle-GO score and the SHT subscale independently predicted successful RTS. In particular, based on Ankle-GO subscales, in both the nonoperative treatment and MBO groups, the SHT showed a significant independent contribution to successful RTS.
Ankle-GO is an objective and comprehensive assessment tool for various criteria related to RTS after nonoperative treatment or surgical treatment.4,18 A systematic review 20 and several studies4,18 reported that Ankle-GO improved in patients with CAI after nonoperative treatment or ankle surgery and that the Ankle-GO score was better in patients who had successful RTS than in those who did not. Our results are consistent with those of previous studies. In this study, the Ankle-GO score was better in patients who achieved successful RTS than in those who did not in the nonoperative treatment and MBO groups. However, in this study, when evaluating the Ankle-GO subscales, we found that in both the nonoperative treatment and MBO groups, SHT and F8T were better in patients who had successful RTS than in those who did not. This is an expected result. In patients with successful RTS, tests such as the SHT and F8T, which involve single-leg hopping, require high ankle stability, as patients must withstand high stresses transmitted to the ankle, and muscle function may also be important.20,25 Therefore, successful RTS may also be related to the recovery or improvement of static (ligamentous) or dynamic (muscular) structures,25,29 which have been consistently reported in patients with CAI or MBO. However, Hardy et al 4 reported a difference in the Ankle-GO score between patients who had successful RTS and non-RTS at 4 months after lateral ankle reconstruction in patients with CAI, but not SHT and F8T. Although the reason for this discrepancy is not fully understood, it may be due to differences in study participants. Picot et al 19 reported that sex may affect successful RTS after a lateral ankle sprain, and female patients are 5 times more likely to have a poor clinical outcome. The participants in the study by Hardy et al included an equal distribution of female and male patients (5:5 ratio), whereas our study did not (3:7 ratio). Therefore, in our study, SHT (11.6 vs 17.9 and 16.0 vs 22.5) and F8T (10.9 vs 18.0 and 14.9 vs 20.3) were superior in both patients with successful and nonsuccessful RTS, respectively, compared with the results of previous studies. Hence, the difference in study participants may have shown different results, and the SHT or F8T may be an important factor in proving the difference between the successful and unsuccessful RTS groups in the nonoperative treatment or MBO groups.
Furthermore, a systematic review 27 and several recent studies3,4 found that ALR-RSI was better in patients with successful RTS than in those without successful RTS after ankle ligament reconstruction. Our findings support previous findings. In this study, only in the MBO group did patients who underwent successful RTS have better ALR-RSI than those who did not. Because the ALR-RSI reflects psychological readiness for RTS after surgery,3,22 it may yield better results in patients with good ankle stability and function after surgery.14,27 Therefore, in nonoperative treatment groups, there may be no difference in the actual structural stability of the ankle, such as ankle ligament healing, and thus there may be no difference in the ALR-RSI between patients with and without successful RTS.
In this study, the SHT subscale of the Ankle-Go was the only subscale to predict successful RTS in either treatment group (nonoperative or MBO). A narrative review 20 reported that the ability to control weightbearing ankle inversion was an important factor for RTS in patients with CAI. The SHT requires quickly and accurately jumping outward and inward between 2 lines spaced 30 cm apart, a movement that is similar to the inversion injury mechanism of the ankle. 29 Therefore, better SHT performance may improve the ability to control ankle inversion, thereby increasing preparation for successful RTS. Thus, among the Ankle-GO subscales, the importance of SHT should be kept in mind for successful RTS decisions.
This study had several limitations. First, this was a retrospective comparative study with a small sample size; a high-quality study with a larger sample size is required. Second, muscle performance was not evaluated. Previous studies9,23,29 reported that muscle performance deficits are important factors affecting ankle function and RTS. Third, we did not evaluate individual sports activities before injury in all the patients participating in this study. This factor may potentially affect the improvement in ankle function after nonoperative treatment or surgery. However, we studied only patients who participated in recreational sports or higher, using a preinjury Tegner activity scale score of ≥6 points. Fourth, we did not assess the compensatory kinetics. Previous studies have found compensatory kinetics in the ankle and hip joints of patients with CAI.17,24 Fifth, there is no way to determine the amount of time and effort spent on rehabilitation in daily life. Nevertheless, to the best of our knowledge, this is the first study to compare the subscales of the Ankle-GO between patients with successful and unsuccessful RTS and based on the Ankle-GO, which is widely used as a criterion for RTS after nonoperative treatment and MBO, and to identify predictive factors for RTS.
Conclusion
Our study demonstrated that in both the nonoperative treatment and MBO groups, SHT and F8T were key factors distinguishing patients with CAI who had successful RTS from those who did not. In addition, SHT was the strongest independent predictor of RTS in both the nonoperative treatment and MBO groups. Therefore, our hypotheses were supported, and clinicians and therapists should consider these findings when planning rehabilitation and RTS strategies.
Footnotes
Final revision submitted December 19, 2025; accepted January 5, 2026.
One or more of the authors has declared the following potential conflict of interest or source of funding: This study was supported by (1) the Korea Medical Device Development Fund, funded by the Korean government (Ministry of Science and ICT; Ministry of Trade, Industry and Energy; Ministry of Health & Welfare; Ministry of Food and Drug Safety; grants KMDF_PR_20200901_0039, KMDF_PR_20200901_0293, RS_2023_00243310); (2) National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (grant NRF-2022R1A2C2092726); (3) Korea University Anam Hospital, Seoul, Republic of Korea (grants K2305161, K2313001, K2312991, I2300231, I2203971); and (4) Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health and Welfare, Republic of Korea (grants RS-2022-KH129293, 2460002156); (5) Clinical Data-Driven Convergent Technology Support Program for Musculoskeletal Human Body Simulation, (Grant No.20220126000001446386). AOSSM checks author disclosures against the Open Payments Database (OPD). AOSSM has not conducted an independent investigation on the OPD and disclaims any liability or responsibility relating thereto. The study protocol was approved by the Institutional Review Board of Korea University Anam Hospital (ED17143).
Availability of Data and Materials
The data that support the findings of this study are available from Jin Hyuck Lee; however, restrictions apply to the availability of these data, which were used under license for the current study and are not publicly available. Furthermore, not all data generated or analyzed during the current study will be disclosed owing to the policies of the Korea University Anam Hospital Research Ethics Board.
