Construction and Validation of a Risk Prediction Nomogram Model for Different Sites of Ankle Pain in Patients with Chronic Ankle Instability Using the OpenSim Environment

doi:10.21203/rs.3.rs-4341654/v1

Download PDF

Article

Construction and Validation of a Risk Prediction Nomogram Model for Different Sites of Ankle Pain in Patients with Chronic Ankle Instability Using the OpenSim Environment

https://doi.org/10.21203/rs.3.rs-4341654/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 27 Sep, 2024

Read the published version in Scientific Reports →

You are reading this latest preprint version

This study aimed to establish a risk prediction nomogram model for anterolateral, mediolateral, and posterolateral ankle pain in patients with chronic ankle instability (CAI) and analyse the risk factors for pain at different ankle sites. Thirty runners with CAI who reported ankle pain in the anterolateral, mediolateral, or posterolateral regions were recruited for this study. Kinematic, kinetic, and electromyographic data during running were collected using motion capture system, etc. These data were used to generate a dynamic nomogram. The results showed that anterolateral ankle pain in patients with CAI may be caused by insufficient gastrocnemius muscle strength (OR = 0.85, 95% CI: 0.73 ~ 0.97), excessive ground reaction force (GRF, OR = 2.64, 95% CI: 1.25 ~ 6.22), and an increased percentage of ankle energy absorption (OR = 9.11, 95% CI: 1.50 ~ 77.79). Mediolateral ankle pain might be contributed by greater ankle inversion angle (OR = 1.08, 95% CI: 1.01 ~ 1.00) and GRF (OR = 2.13, 95%CI: 1.17 ~ 4.31). Moreover, posterolateral ankle pain was predicted by increased ankle internal rotation angle (OR = 1.06, 95% CI: 1.00-1.12), increased GRF (OR = 2.16, 95% CI: 1.07–4.80), and decreased dynamic stability (OR = 0.20, 95% CI: 0.05–0.68). To prevent ankle pain, patients with CAI should be encouraged to focus on improving the neuroreceptor sensitivity of the gastrocnemius muscles, and retraining their energy absorption patterns.

Health sciences/Risk factors

Health sciences/Health care/Disease prevention

Health sciences/Health care/Medical ethics

CAI

ankle pain

risk prediction nomogram model

OpenSim model

Patients with CAI exhibit a heightened susceptibility to developing lateral ankle pain while engaging in running activities.
Lateral ankle pain can significantly hinder runners' performance and predispose them to compensatory injuries.
The precise biomechanical mechanisms underlying lateral ankle pain are currently ambiguous, posing challenges for targeted prevention strategies.
Furthermore, there is a lack of comprehensive understanding regarding the biomechanical mechanisms contributing to ankle pain in distinct regions.

WHAT THIS STUDY ADDS

This study employed a risk prediction nomogram model to elucidate the biomechanical mechanisms underlying anterolateral, mediolateral, and posterolateral ankle pain.
Anterolateral ankle pain in patients with CAI is caused by insufficient gastrocnemius muscle strength, greater GRF, and an increased percentage of energy absorption in the ankle joint during running.
Mediolateral ankle pain was found to be associated with a higher ankle inversion angle and GRF.
Posterolateral ankle pain is predicted by an elevated ankle internal rotation angle, increased GRF, and decreased dynamic stability.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

Upon elucidating the distinct biomechanical mechanisms of lateral ankle pain across various regions, clinicians can tailor intervention strategies according to the specific location of the patient's pain. Furthermore, future research endeavors may focus on conducting randomized controlled trials with patients to validate the efficacy of targeted preventive measures for ankle pain.

Running is a highly popular sport due to its accessibility. It requires no equipment, playgrounds, or time constraints and has minimal technical requirements. Therefore, it naturally becomes the preferred exercise option for sports enthusiasts. The 2022 Research on National Fitness Behaviour and Consumption, released by the China Sporting Goods Federation, revealed that running had the highest participation rate among sports, accounting for 61.0% of participants. Nonetheless, similar to other forms of physical exercise, participation in running poses considerable risks of musculoskeletal injuries. A survey showed that approximately 40–50% of runners suffered from running-related injuries, with lateral ankle sprains being the most common ¹. Additionally, forty percent of patients with lateral ankle sprains would develop CAI ². This population commonly experiences repetitive ankle injuries, proprioceptive deficits, instances of “giving way”, and ankle pain during exercise, which severely affects their quality of life and sports participation ².

Ankle pain is the most common symptom in patients with CAI. According to a systematic review, approximately 60.1% of patients with CAI experienced severe lateral ankle pain, with 47.7% of cases occurring during moderate- and high-intensity physical activity ³. Indeed, ankle pain could profoundly disrupt a runner's movement patterns, leading to a plethora of compensatory injuries. First, to maintain a stable movement pattern under external stimuli, humans must process complex somatosensory information while running ⁴. However, sensations of ankle pain might project to various areas of the brain, including the somatic cortex, limbic system, and frontal cortex. This could potentially affect proprioception at multiple levels in the central nervous system, may be detrimental to maintaining running stability, reduces the effectiveness of exercise, and trigger compensatory injuries ⁴. Second, Bouffard et al. ⁵ conducted a comprehensive biomechanical analysis to scrutinize the alterations in movement patterns induced by ankle pain. Their findings elucidated that individuals with lateral ankle pain demonstrated compromised activation of the gluteus medius muscle and restricted activation of the gastrocnemius muscle. This type of movement pattern would reduce the efficiency of energy absorption in the extensor muscles of the lower extremity, resulting in the affected leg being exposed to larger GRF over an extended period. Third, recent research indicated that up to 55% of cases of knee osteoarthritis are associated with lateral ankle pain ⁶. Furthermore, a cross-sectional study revealed that, unlike controls, individuals with lateral ankle pain exhibited an extended stance phase, diminished stride length, increased ankle plantarflexion angle, and restricted hip flexion movements while engaged in running ⁷. In the long term, the above movement strategy could damage the arch and the soft tissue structures around the ankle joint ⁷.

Due to the severe negative impact of lateral ankle pain, the implementation of early intervention training has been recommended for populations that are predisposed to ankle pain, especially individuals with CAI. The identification of biomechanical risk factors is an indispensable prerequisite for targeted intervention training. However, the biomechanical risk factors for ankle pain during running in patients with CAI remain unknown. Concurrently, the absence of relevant biomechanical studies has precluded clinicians from developing effective and targeted prevention strategies. Furthermore, several studies have highlighted that ankle pain could occur at different sites, including the anterolateral, mediolateral, and posterolateral parts of the ankle, and that the biomechanical risk factors associated with ankle pain exhibit potential modification contingent upon the precise anatomical site where the pain is localized ^8,9,10. Therefore, to further our current understanding of the biomechanical risk factors associated with ankle pain at different sites, it was necessary to develop biomechanical risk prediction models based on different sites of ankle pain. Recently, risk predictive models, such as logistic regression, nomograms, and machine learning, have been increasingly used for diagnostic and clinical prognosis of specific diseases ¹¹. The risk prediction nomogram model was instrumental in preventing and rehabilitating gastric cancer, colorectal cancer, and even COVID-19 after its predictive efficacy was confirmed by scientific methods. This method also offers new ideas for establishing risk prediction models for ankle pain ¹¹.

This study aimed to investigate the biomechanical risk factors associated with ankle pain at different locations and to develop a risk prediction nomogram model for anterolateral, mediolateral, and posterolateral ankle pain in patients with CAI to provide a theoretical basis for the prevention and early management of ankle pain. Based on previous studies, this research hypothesized that anterolateral and mediolateral ankle pain experienced by patients with CAI might be associated with inadequate gastrocnemius muscle strength, excessive ankle inversion angle, higher ankle energy absorption, and GRF. In addition, greater ankle internal rotation angle, GRF, and poorer dynamic stability might be risk factors for posterolateral ankle pain.

Participants

This study analysed male runners with CAI who ran at least 10 km weekly. The diagnosis of CAI was based on the criteria of the International Ankle Society. Furthermore, the study’s inclusion criteria included the following: 1) Participants that had a history of at least one ankle sprain that caused a 1-day interruption in physical activity. The first ankle sprain must have occurred more than 1 year prior to the experiment, and the most recent ankle sprain must have occurred more than 3 months prior to the experiment. 2) The patient should have experienced two or more instances of severe 'giving way' in the 6 months prior to the test, as well as having had two or more ankle sprains during that period. Additionally, the Cumberland Ankle Instability Questionnaire (CAIT) score of the included participants had to be less than 24. 3) A Foot and Ankle Ability Measure (FAAM) of less than 80% and a Foot and Ankle Outcome Score (FAOS) of less than 75%. The exclusion criteria were as follows: 1) Patients who had undergone bone, muscle, or joint surgery; 2) Patients with a history of lower extremity fractures; 3) Patients who experienced acute lower extremity injuries within 3 months prior to the experiment, which altered the integrity and function of the joint and interrupted physical activity for > 1 day.

The study included 123 participants, who were selected based on the aforementioned criteria. During the pre-test survey, three participants were excluded due to knee surgery experience, leaving 120 participants in the experiment. The participants were divided into four groups: the anterolateral ankle joint pain, mediolateral pain, posterolateral pain, and no pain groups. All participants were required to carefully read the experiment protocol and sign informed consent forms. The study received approval from the Ethics Committee of East China Normal University. Before the formal experiment, the participants underwent basic anthropometric tests, including height, weight, and CAIT score. The results are shown in Table 1.

Table 1

Demographics of the participants (mean ± SD)
Location of ankle pain	sample size (n)	height (cm)	weight (kg)	CAIT score
No pain	30	178.9 ± 5.2	67.4 ± 6.5	22.1 ± 1.7
Anterolateral	30	179.9 ± 6.8	68.1 ± 5.9	20.4 ± 2.2
Mediolateral	30	177.2 ± 5.8	67.9 ± 6.5	21.4 ± 2.1
Posterolateral	30	178.7 ± 5.2	69.3 ± 5.1	20.5 ± 2.6

Equity, diversity, and inclusion statement

Our study included all identified cases of ankle pain in college student-athletes inclusive of all race/ethnicities, socioeconomic levels, and occurrence in a marginalized community. Our author team consisted of one woman and four men from different disciplines (athletic training, biomechanics), including two authors considered junior scholars. The data collection was similar in all cases and we did not alter methods based on regional, educational, or socioeconomic differences of the community in which the case occurred.

Test Apparatus

Motion capture system

In this study, 12 Vicon Vero2.2 (Vicon Industries, Inc., USA) infrared cameras and a motion capture system with a 14 mm infrared marker ball were used to record single-leg landing manoeuvres. The cameras were positioned at a height of 2.5 m with a sampling frequency of 100 Hz.

Three-dimensional force plate

In this study, two 40 cm × 60 cm Kistler 3D force plates (Kistler Inc., Switzerland) were used to record the GRF of the participants during single-leg landing with a sampling frequency of 1 kHz.

Surface EMG testing system

The surface EMG equipment used in this study was the Telemyo 2400DTS sEMG wireless telemetry system (Noraxon Company, USA), which contains a 32-channel signal collector, a USB data transfer interface, and a wireless receiver card and could be connected to a computer either wired or wirelessly. For surface EMG testing, the equipment is often used in conjunction with electrodes and pre-amplifiers.

Experimental Design and Testing Procedures

The study was conducted on October 15, 2023, at the key laboratory of adolescent health assessment and exercise intervention of Ministry of Education, East China Normal University. Prior to the test, all participants were provided with running shoes and were asked to jog on a treadmill at a speed of 3.5 m/s for 10 min to warm up. The testers gave the participants detailed explanations of the test movements and requirements. Following this, the participants were affixed with reflective markers and electrodes. The markers were meticulously affixed to the relevant bony landmarks based on the Plug In Gait-full body model, while the electrodes were securely positioned on the muscle bellies of the peroneus longus, gastrocnemius, and tibialis anterior muscles. Before the formal experiment, the participants underwent static calibration, which involved standing naturally with a straight torso, arms abducted, elbows flexed at 90°, and palms pronated while looking straight ahead. Simultaneously, the maximal voluntary contraction (MVC) of the muscles, including those of the peroneus longus, gastrocnemius, and tibialis anterior muscles, were measured as the basis for the standardization of EMG signals. At the beginning of the experiment, the participants were asked to run on a 15 × 2 m treadmill at a comfortable speed with a movement pattern similar to that of daily exercise, as shown in Fig. 1. Two three-dimensional force platforms were embedded in the middle of the treadmill. The participants were instructed to accurately measure their step lengths to ensure that the affected leg could strike the force platform accurately while running. Overall, five valid sets of data were collected, with a 3-min interval between each test.

Data Extraction and Preparation

Kinematics

The kinematic data were calculated using Visual 3D software (C-Motion Inc., USA). The marker’s trajectory was filtered using a 10-Hz, four-step Butterworth filter. Afterwards, the joint angles in the sagittal, frontal, and horizontal planes of the ankle joint were calculated at the moment of initial contact. The calculations included building a transformation matrix, calculating the joint centre, and establishing a local coordinate system for the human body segment. The kinematic parameters analysed in this study focused on the following: 1) Ankle plantarflexion and dorsiflexion angle: in the present study, the angle was defined as 0° when the foot was vertical to the shank, and on that basis, the dorsiflexion angle was positive and the plantarflexion angle was negative; 2) ankle inversion and eversion angle, which was set at 0° in the neutral position of the ankle joint, positive for inversion and negative for eversion; and 3) the internal and external rotation angle of the ankle: the angle was determined as 0° when the ankle was in a neutral position in the horizontal plane. It was positive during internal rotation and negative during external rotation, as shown in Fig. 2.

Kinetics

In this study, we focused on analysing kinetic data during the stance phase of the affected leg in patients with CAI. Raw GRF data were subjected to filtration utilizing a four-step Butterworth low-pass filter with a cut-off frequency of 50 Hz. In this research investigation, the running stance period was defined upon the evaluation of the GRF. Specifically, if the GRF exceeded 20 N, it unequivocally indicated that the participants were undergoing the moment of initial contact. Conversely, if the GRF registered below the threshold of 20 N, it reliably indicated that the participant was engaging in the pivotal action of taking off. The core indicators analysed in this study include vertical GRF, TTP, dynamic stability, and ankle energy absorption.

TTP was calculated using the following formula:

$$\frac{{GRF}_{Peak}}{{t}_{Peak}-{t}_{initial contact}}$$

Where t is time measured in ms.

(2) Dynamic stability was obtained by calculating the displacement of the centre of pressure (COP) during the stance phase.

(3) The formula for ankle energy absorption was as follows

$$\text{a}\text{n}\text{k}\text{l}\text{e} \text{e}\text{n}\text{e}\text{r}\text{g}\text{y} \text{a}\text{b}\text{s}\text{o}\text{r}\text{p}\text{t}\text{i}\text{o}\text{n}=\int {P}_{minus value}\left(t\right)dt$$

Where P is the ankle power, which could be calculated by multiplying the ankle moment by the angular velocity, and t is time.

Muscle force

(1) OpenSim Modelling

The Marks trajectory and GRF data in c3d format files were imported into Matlab 2016b for conversion. The Marks trajectory data was written into trc files, and the GRF data was written into mot files. OpenSim was used to load the trc and mot files. First, to ensure the mass and length of the human body segment in the model corresponded to real-life values, the data was scaled using the Scaling Model (SM) module. Second, inverse dynamics were used to calculate “the optimal match point” between the model and reality by solving the optimization problem of weighted least squares that minimized the marking error. Model assumptions, kinematic system errors, and noise occasionally caused dynamic inconsistencies in the model. To address this, OpenSim used the Reduce Residuals Algorithm (RRA) module to minimize the difference between the experimental results and the model simulation results (F_residual). Finally, the Calculate Muscle Control (CMC) module was then utilized to simulate lower extremity muscle force, as shown in Fig. 3.

(2) Model Validation

The model validation process included three aspects: kinematic validation, kinetic validation, and electromyographic validation. To validate kinematics, we ensured that the disparity between the experimentally measured peak joint angle and the peak joint angle simulated by OpenSim did not exceed 5°. For kinetics, a model was considered to adequately fit the actual movement if the difference between the experimentally measured peak pelvic moments (or residual forces) and the results of the OpenSim simulation was less than 75 Nm and 20 N, respectively. As for electromyography, we compared the experimental results with the electromyographic curves obtained from modelling. If the results were similar, we considered the results of RRA and CMC to be acceptable.

Statistics

Statistical analysis and the risk prediction nomogram model were performed using SPSS version 26.0 (IBM, USA) and R Studio 4.3.1 (R studio, USA). The normality of the data was tested using the Shapiro-Wilk test. One-way ANOVA was used to compare the biomechanical characteristics of different ankle pain locations (no pain, anterolateral, mediolateral, and posterolateral). The results are presented as means ± SD. Lasso regression was used to select potential risk factors associated with ankle pain, build logistic regression models, and generate a dynamic nomogram for risk prediction in R Studio. The predictive efficacy, specificity, computational sensitivity, and predictive gain of the model were verified by plotting ROC curves, calibration curves, and DCA curves. All statistical tests in this study were considered statistically significant at P < 0.05.

The results of one-way ANOVA (Fig. 4) showed that, compared to patients with CAI without pain, patients with CAI who had anterolateral ankle pain had lower gastrocnemius muscle force during running (F_[1,118] = 6.198, P = 0.001). Additionally, patients with CAI who had mediolateral ankle pain exhibited a significantly greater ankle inversion angle during running compared to patients with CAI without pain (F_[1,118] = 4.772, P = 0.001).

Patients with CAI who had anterolateral ankle pain exhibited a greater GRF during running than did patients with CAI without pain (F_[1,118] = 7.347, P = 0.002). Additionally, ankle energy absorption was significantly greater in patients with CAI who had anterolateral ankle pain during running compared to patients with CAI without pain (F_[1,118] = 3.451, P = 0.002), as shown in Fig. 5.

As illustrated in Fig. 6, the study employed LASSO regression to select potential risk factors, which were then used as independent variables in logistic regression. The study identified four risk factors associated with anterolateral ankle pain: gastrocnemius muscle force, ankle internal rotation angle, GRF, and ankle energy absorption. The correlation coefficients for these factors were − 0.106, -0.001, 0.652, and 1.308, respectively. Additionally, in the context of mediolateral ankle pain, our investigation identified three significant risk factors: peroneus longus muscle force, ankle internal rotation angle, and GRF. The correlation coefficients for these variables were found to be -0.003, -0.030, and − 0.359, respectively. The risk factors associated with posterolateral ankle pain were peroneus longus muscle force, gastrocnemius muscle force, ankle internal rotation angle, GRF, dynamic stability, and ankle energy absorption. The correlation coefficients for these risk factors were − 0.022, -0.025, -0.033, -0.395, -0.819, and − 0.235, respectively.

As indicated in Table 2, the variables that passed the screening process were used in the logistic regression model. The results also indicated that gastrocnemius muscle force (OR = 0.85, 95%CI: 0.73 ~ 0.97), GRF (OR = 2.64, 95%CI: 1.25 ~ 6.22), and ankle energy absorption (OR = 9.11, 95%CI: 1.50 ~ 77.79) were independent predictors of anterolateral ankle pain. Additionally, ankle inversion angle (OR = 1.08, 95% CI: 1.01–1.18) and GRF (OR = 2.13, 95% CI: 1.17–4.31) were found to be independent predictors of mediolateral ankle pain. Furthermore, ankle internal rotation angle (OR = 1.06, 95% CI: 1.00 ~ 1.12), GRF (OR = 2.16, 95% CI: 1.07 ~ 4.80), and dynamic stability (OR = 0.20, 95% CI: 0.05 ~ 0.68) were independent predictors of posterolateral ankle pain.

Table 2

Results of Stepwise Logistic Regression
Variables	Anterolateral ankle pain		Mediolateral ankle pain		Posterolateral ankle pain
Variables	OR (95%CI)	P value	OR (95%CI)	P value	OR (95%CI)	P value
peroneus longus muscle force
Gastrocnemius muscle force	0.85 (0.73, 0.97)	0.02
Tibialis anterior muscle force
Ankle plantarflexion angle
Ankle inversion angle			1.08 (1.01, 1.18)	0.05
Ankle internal rotation Angle					1.06 (1.00, 1.12)	0.04
GRF	2.64 (1.25, 6.22)	0.02	2.13 (1.17, 4.31)	0.02	2.16 (1.07, 4.80)	0.04
TTP
COP displacements					0.20 (0.05, 0.68)	0.02
Ankle energy absorption	9.11 (1.50, 77.79)	0.02

A risk prediction nomogram model of ankle pain in CAI was established based on the results of logistic regression, which was shown in Figs. 7–9. In the nomogram, a higher total score indicated a higher risk of ankle pain. Figure 6 shows the risk prediction nomogram model of participants with anterolateral ankle pain. Using the fifth observed value as an example, the participant’s ankle energy absorption was 1.25J·kg^− 1, corresponding to a score of 1.06. The GRF was 6.6BW, corresponding to a score of 1.5. The gastrocnemius muscle force was 7.3BW, corresponding to a score of 0.25. The total score of the three indexes above was 2.81. Therefore, the likelihood of experiencing anterolateral ankle pain in patients with CAI at this time was 94.4%.

Figure 8 shows the risk prediction nomogram model for participants with mediolateral ankle pain, which included two predictors: GRF and ankle inversion angle. For instance, in the 11th observed value of this experiment, the participant’s GRF during running was 6.9 BW, corresponding to a score of 1.88, and the ankle inversion angle was 14°, corresponding to a score of 0.78. The total score of the two indicators was 2.68, indicating a 93.8% probability of mediolateral ankle pain.

The risk prediction nomogram model for posterolateral ankle pain comprised three predictors: dynamic stability, GRF, and ankle internal rotation angle. Using the 15th observed value of this experiment as an example, the participant exhibited a COP displacement of 3.02 during running, which corresponds to a score of 0.85. Additionally, the participant had a GRF of 4.05 BW, corresponding to a score of -0.05, and an ankle internal rotation angle of 11°, corresponding to a score of 0.55. The combined score for the three indexes was 1.35. At this specific instance, the probability of posterior lateral ankle pain was 78.6%.

The study examined the risk prediction nomogram model by plotting the ROC, calibration, and DCA curves. The ROC curve assessed the model’s ability to distinguish between non-events and events, and its discrimination ability was reflected by the area under the curve (AUC). The AUCs of the nomogram in this study were 0.85, 0.77, and 0.76, respectively, indicating that the model had excellent distinctive ability. The calibration curves measured the degree of correspondence between the predicted and actual results. This was accomplished by resampling the training set 1000 times using the bootstrap method. The calibration curves indicated that the deviation of the actual curves from the ideal line was small in this study, suggesting that the risk prediction nomogram model was well calibrated. The DCA curves indicated that the risk prediction nomogram model could offer clinical benefits to runners with CAI who had anterolateral ankle pain and were at risk thresholds ranging from 0 to 0.90, provide clinical benefits to runners with CAI with mediolateral ankle pain at risk thresholds ranging from 0 to 0.99, and give clinical benefits to runners with CAI with posterolateral ankle pain at risk thresholds ranging from 0 to 0.9. Overall, the model's predictions were moderately accurate, as shown in Fig. 10.

The gastrocnemius muscle—one of the ankle plantar flexors—is connected to the foot through the Achilles tendon. During physical activity, the dynamic contraction of the gastrocnemius muscle plays a crucial role in generating an ankle plantarflexion moment. This enhances pedal efficiency and stretching capabilities, and mitigates the angular velocity of ankle dorsiflexion. Consequently, it significantly reduces the sagittal plane load exerted on the ankle joint. Moreover, this mechanism further bolsters ankle stability during the stance phase ¹². Patients with CAI often have reduced gastrocnemius muscle force during running due to multiple ankle sprains, which could lead to insufficient ankle muscle proprioception and inadequate neural control ¹³. In this study, gastrocnemius muscle force insufficiency was found to be one of the risk factors for anterolateral ankle pain in patients with CAI, which was consistent with the findings of previous studies ¹⁴. Anderson et al. ¹⁴ found that gastrocnemius muscle force insufficiency was present in approximately 88% of patients with anterolateral ankle pain. Fourchet et al. ¹⁵ found that after inducing gastrocnemius fatigue, runners with inadequate gastrocnemius muscle force often displayed longer forefoot stance time during running. The intrinsic mechanism was that gastrocnemius muscle weakness reduced the pedal and stretch force during running, and runners strategically extended the duration of forefoot support in order to maintain the impulse generated during the swing phase ¹⁵. However, prolonged stance time could lead to arch collapse, which in the long term manifests as anterolateral ankle pain ¹⁵. Therefore, patients with CAI may benefit from performing plyometric training to improve the sensitivity of gastrocnemius neuroreceptors (e.g., muscle spindle, Golgi apparatus, and tendon organs), which may improve neuromuscular control around the gastrocnemius muscle and increase gastrocnemius strength ¹⁶.

Patients with CAI often exhibit a higher ankle inversion angle during movement due to impaired ankle inversion proprioception ¹⁷. The findings of this study suggest that a greater ankle inversion angle is associated with mediolateral ankle pain. This might be because repeated ankle inversion could accelerate the formation of scar tissue and alter collagen types, water content, and proteoglycan abundance in the lateral ankle ligaments. These changes reduce the ligament's viscoelasticity and often lead to pain during ankle inversion ¹⁸. Furthermore, Park et al. ¹⁹ highlighted that repeated excessive ankle inversion might lead to frequent collisions between the talus and distal tibia, which could result in ankle impingement syndrome. Ankle impingement syndrome has also been associated with mediolateral ankle pain. Moreover, repeated collisions between the talus and tibia could result in insufficient blood supply to the talus, leading to inadequate healing of cartilage contusions and necrosis of cartilage debris, which could ultimately cause mediolateral ankle pain. Therefore, runners with CAI should strengthen their ankle evertors to provide sufficient eversion moments during the stance phase. Rehabilitative attention should also be given to proprioceptive training of the ankle joint to maintain a neutral position at the initial contact moment and reduce mediolateral ankle pain.

Research has shown that patients with CAI often have an excessive ankle internal rotation angle during running ²⁰. This could be attributed to the longer activation time of the ankle evertors in these patients compared to healthy individuals, which limits the ability of the muscle groups to decrease the ankle internal rotation angle ²¹. Furthermore, patients with CAI are accustomed to placing the COP on the lateral side of the foot, which would lead to a higher ankle internal rotation angle ²². According to the finding of Wu et al. ²³, an ankle internal rotation moment of up to 10 Nm could cause ankle pain and discomfort, moments of 23 Nm may result in a grade I ankle sprain and those of 40–45 Nm or more would lead to an ankle fracture. Kobayashi et al. ²⁰ further discovered that excessive ankle internal rotation, ranging from 5°–7°, had the potential to induce an increase in anterior-posterior displacement of the talocrural joint, stress on the posterior talofibular ligament, and may lead to ankle pain.

This study's findings are consistent with those of previous research, as the prediction model indicates that the ankle internal rotation angle is one of the predictors of posterolateral ankle pain. Therefore, the study’s findings indicate that patients with CAI may benefit from engaging in jump stabilization training, which increases the strength of the ankle evertors, to enhance body balance. Moreover, patients with CAI could focus on optimizing the COP position during running to reduce excessive ankle internal rotation angle. Ultimately, these interventions would prevent the manifestation of posterolateral ankle pain ²⁴.

Patients with CAI tend to display greater GRF during running. This is caused by the disruption of the structure of the anterior talofibular ligament due to grade II or III ankle sprains, which could lead to the formation of scar tissue during the ligament-remodelling phase. The scar tissue in the damaged ligament is less mobile than the intact ligament, and may lead to anterior movement of the talus, which reduces the range of motion (ROM) during ankle dorsiflexion ²⁵. However, studies have consistently shown that the ankle dorsiflexion ROM during running should be at least 20°~30°²⁵. Restricted ankle dorsiflexion has been demonstrated to strongly correlate with the occurrence of excessive GRF within the ankle joint during the stance phase. The study's prediction model found that greater GRF during the stance phase was a risk factor for anterolateral, mediolateral, and posterolateral ankle pain. A higher GRF cannot be adequately absorbed by ankle sagittal plane motion alone; therefore, ankle frontal plane motion is required to increase the displacement distances in the medial-lateral direction of the COP during running. This energy absorption strategy aggravated the lateral load on the ankle joint. Furthermore, an inappropriate positioning of the COP could lead to the impingement of the distal tibia on the talus. Persistent impingement irritation would consequently lead to the formation of soft tissue scarring and the development of bone capillaries. Subsequently, ankle swelling and pain could ensue as a result of these processes ²⁵. The findings of this study and those of previous research indicate that patients with CAI should focus on retraining their running posture. The forefoot strike strategy might also be effective in cushioning GRF and preventing lateral ankle pain ²⁶.

Patients with CAI frequently exhibit inadequate dynamic stability while running. This was due to ankle instability that affects not only the activation pattern of muscles around the ankle joint but also the neuromuscular control of the hip joint, which is mainly characterized by a decrease in hip extensor muscle strength, excessive displacement range of centre of mass (COM), and poorer dynamic stability ²⁷. Furthermore, an in vivo study revealed that during running, talus displacement was greater in patients with CAI than in healthy individuals due to ligament laxity, which may also result in a loss of dynamic stability ²⁸. Predictive modelling results indicate that poorer dynamic stability in patients with CAI was associated with posterolateral ankle pain. Furthermore, insufficient dynamic stability increased the total ankle displacement during running in patients with CAI, which repeatedly eroded and squeezed soft tissues and ligaments around the ankle, leading to local inflammation and ultimately ankle pain ²⁹. To improve dynamic stability, ankle braces are commonly used to provide mechanical support and restore the motion pattern to pre-injury states in patients with CAI ³⁰. Meanwhile, strength training of the hip extensors would improve core control during running and indirectly decrease the risk of ankle pain.

During the stance phase of running, the human body must absorb energy, primarily using the ankle plantar flexors, knee extensors, and hip extensors. Delahunt et al. ³¹ reported that the peak of vertical and anterior-posterior GRF occurred in the early stage of the stance phase. The musculoskeletal structures of the lower extremity ensure complete energy absorption within a short time. However, due to the presence of sensory-motor deficits in patients with CAI, they cannot adequately allocate the proportion of joint energy absorption ³², and this exposes the ankle joint to heightened energy absorption load ³². The prediction model in this study revealed that excessive energy absorption in the ankle joint could cause anterolateral ankle pain. This may be because the ankle joint requires significant plantarflexion and dorsiflexion activity for complete energy absorption. Furthermore, during ankle plantarflexion, there was an augmented burden on the anterior talofibular ligament, leading to amplified stress and strain. Conversely, dorsiflexion of the ankle joint intensifies the stress and strain on the calcaneofibular ligament. Repeated overstretching could damage the ligament’s capillaries, leading to haemorrhage into the surrounding joint capsule and pain. Consequently, if a lateral force was applied to the ankle joint, the level of pain experienced by these individuals might increase significantly ³³. Therefore, individuals with CAI should focus on retraining their energy absorption strategies to prevent lateral ankle pain.

Clinical Implications

This study, utilizing the risk prediction nomogram model, has unveiled the biomechanical risk factors associated with ankle pain in various regions. Individuals with anterolateral ankle pain should focus on strengthening the gastrocnemius muscle through heel raise exercises and adjusting their lower extremity cushioning method while running to decrease ankle energy absorption. For those with mediolateral pain related to chronic ankle instability (CAI), enhancing the strength of ankle evertor muscles, like the peroneus longus muscle, is crucial to prevent excessive ankle inversion during landing. Patients experiencing posterolateral pain should prioritize improving proprioception and ankle joint control to prevent internal rotation during landing. It is worth noting that CAI patients across all pain locations exhibit higher ground reaction forces during landing, underscoring the importance of modifying cushioning patterns during running to land with the lower limbs in a more flexed position.

Limitations

In this study, a risk prediction nomogram model was used to analyse the risk factors for anterolateral, mediolateral, and posterolateral ankle pain; however, the study has some limitations. First, this study was unable to distinguish the degree of ankle pain due to the limited sample size. Future studies could explore the biomechanical risk factors of different degrees of ankle pain using the visual analogue scale (VAS). Second, this study solely analysed the contribution of ankle biomechanical characteristics to ankle pain. Future studies could consider including relevant indicators of the proximal joint.

This study showed that anterolateral ankle pain in patients with CAI is caused by insufficient gastrocnemius muscle strength, greater GRF, and an increased percentage of energy absorption in the ankle joint during running. Furthermore, mediolateral ankle pain was found to be associated with a higher ankle inversion angle and GRF, while posterolateral ankle pain was predicted by an elevated ankle internal rotation angle, increased GRF, and decreased dynamic stability. To prevent ankle pain, patients with CAI should focus on improving the neuroreceptor sensitivity of the gastrocnemius muscles, supplementing proprioceptive training of ankle evertor, balance training, and retraining their energy absorption strategy.

CAI chronic ankle instability

ROC receiver operating characteristic

DCA decision curve analysis

AUC respective area under the curve

GRF ground reaction force

CAIT Cumberland Ankle Instability Questionnaire

FAAM Foot and Ankle Ability Measure

FAOS Foot and Ankle Outcome Score

ROM range of motion

VAS visual analogue scale

Ethics approval and consent to participate

The study was performed in accordance with the ethical standards of the Declaration of Helsinki given ethics approval was obtained from the Ethics Committee of East China Normal University on 11 December 2023. Furthermore, all participants were required to carefully read the experiment protocol and sign informed consent forms.

Consent to Participate

Consent for the publication of personal data in this study was obtained from the participants.

Data Availability Statement

Data is provided within the manuscript or supplementary information files

Competing Interests

The authors have no competing interests or conflicts with the content of this article to declare.

Funding

The authors have not received a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

Author’s Contributions

According to the definition given by the International Committee of Medical Journal Editors (ICMJE), the authors listed qualify for authorship based on making one or more substantial contributions to the intellectual content of the manuscript. Conceptualisation: Zeyi ZHANG, Youping SUN. Ethics: Zeyi ZHANG, Youping SUN. Funding acquisition: Youping SUN. Methodology: Zeyi ZHANG, Jian SONG, Hanlin SHI, Shengmeng WEI. Data collection: Zeyi ZHANG, Jian SONG, Hanlin SHI, Shengmeng WEI. Data cleaning: Zeyi ZHANG, Jian SONG,. Data analysis: Zeyi ZHANG, Hanlin SHI. Data interpretation: Zeyi ZHANG, Shengmeng WEI. Writing—original draft preparation: Zeyi ZHANG. Writing—review and editing: Youping SUN. Project administration: Youping SUN. All authors have read and agreed to the published version of the manuscript.

Acknowledgements

The authors express their gratitude to the Biomechanics Laboratory of East China Normal University for providing equipment support and to all study participants for their valuable contributions.

Author’s Information

N/A.

Patient and Public Involvement

Patients or the public were not involved in the design, or conduct, or reporting, or dissemination plans of our research.

COLAPIETRO M, FRASER J J, RESCH J E, et al.Running mechanics during 1600 meter track runs in young adults with and without chronic ankle instability[J]. Phys Ther Sport,2020,42:16-25.
STOTZ A, JOHN C, GMACHOWSKI J, et al.Effects of elastic ankle support on running ankle kinematics in individuals with chronic ankle instability and healthy controls[J]. Gait Posture,2021,87:149-155.
ADAL S A, MACKEY M, POURKAZEMI F, et al.The relationship between pain and associated characteristics of chronic ankle instability: A retrospective study[J]. J Sport Health Sci,2020,9(1):96-101.
BERTRAND-CHARETTE M, LE QUANG M, ROY J S, et al.Alteration of ankle proprioceptive threshold during gait in the presence of acute experimental pain[J]. PLoS One,2022,17(1):e0263161.
BOUFFARD J, BOUYER L J, ROY J S, et al.Tonic pain experienced during locomotor training impairs retention despite normal performance during acquisition[J]. J Neurosci,2014,34(28):9190-9195.
PATERSON K L, HINMAN R S, HUNTER D J, et al.Impact of concurrent foot pain on health and functional status in people with knee osteoarthritis: data from the osteoarthritis initiative[J]. Arthritis Care Res (Hoboken),2015,67(7):989-995.
KOLDENHOVEN R M, VIROSTEK A, DEJONG A F, et al.Increased Contact Time and Strength Deficits in Runners With Exercise-Related Lower Leg Pain[J]. J Athl Train,2020,55(12):1247-1254.
VEGA J, PEñA F, GOLANó P.Minor or occult ankle instability as a cause of anterolateral pain after ankle sprain[J]. Knee Surg Sports Traumatol Arthrosc,2016,24(4):1116-1123.
BROWN G P, FEEHERY R V, JR., GRANT S M.Case study: the painful os trigonum syndrome[J]. J Orthop Sports Phys Ther,1995,22(1):22-25.
KOSHINO Y, WATANABE K, AKIMOTO M, et al.Factors associated with persistent pain in college athletes with a history of lateral ankle sprain[J]. Phys Ther Sport,2023,64:27-31.
XIE W, LI Y, MENG X, et al.Machine learning prediction models and nomogram to predict the risk of in-hospital death for severe DKA: A clinical study based on MIMIC-IV, eICU databases, and a college hospital ICU[J]. Int J Med Inform,2023,174:105049.
PASCUAL HUERTA J.The effect of the gastrocnemius on the plantar fascia[J]. Foot Ankle Clin,2014,19(4):701-718.
JEON H G, LEE S Y, PARK S E, et al.Ankle Instability Patients Exhibit Altered Muscle Activation of Lower Extremity and Ground Reaction Force during Landing: A Systematic Review and Meta-Analysis[J]. J Sports Sci Med,2021,20(2):373-390.
ANDERSON J G, BOHAY D R, ELLER E B, et al.Gastrocnemius recession[J]. Foot Ankle Clin,2014,19(4):767-786.
FOURCHET F, KELLY L, HOROBEANU C, et al.High-intensity running and plantar-flexor fatigability and plantar-pressure distribution in adolescent runners[J]. J Athl Train,2015,50(2):117-125.
HUANG P Y, JANKAEW A, LIN C F.Effects of Plyometric and Balance Training on Neuromuscular Control of Recreational Athletes with Functional Ankle Instability: A Randomized Controlled Laboratory Study[J]. Int J Environ Res Public Health,2021,18(10).
YU R, YANG Z, WITCHALLS J, et al.Kinesiology tape length and ankle inversion proprioception at step-down landing in individuals with chronic ankle instability[J]. J Sci Med Sport,2021,24(9):894-899.
LIN C Y, SHAU Y W, WANG C L, et al.Modeling and Analysis of the Viscoelastic Response of the Ankle Ligament Complex in Inversion Ankle Sprain[J]. Ann Biomed Eng,2015,43(9):2047-2055.
PARK B S, CHUNG C Y, PARK M S, et al.Inverse Relationship Between Radiographic Lateral Ankle Instability and Osteochondral Lesions of the Talus in Patients With Ankle Inversion Injuries[J]. Foot Ankle Int,2019,40(12):1368-1374.
KOBAYASHI T, SAKA M, SUZUKI E, et al.In vivo kinematics of the talocrural and subtalar joints during weightbearing ankle rotation in chronic ankle instability[J]. Foot Ankle Spec,2014,7(1):13-19.
SIMPSON J D, KOLDENHOVEN R M, WILSON S J, et al.Ankle kinematics, center of pressure progression, and lower extremity muscle activity during a side-cutting task in participants with and without chronic ankle instability[J]. J Electromyogr Kinesiol,2020,54:102454.
HUANG P Y, LIN C F, KUO L C, et al.Foot pressure and center of pressure in athletes with ankle instability during lateral shuffling and running gait[J]. Scand J Med Sci Sports,2011,21(6):e461-467.
WEI F, FONG D T, CHAN K M, et al.Estimation of ligament strains and joint moments in the ankle during a supination sprain injury[J]. Comput Methods Biomech Biomed Engin,2015,18(3):243-248.
HALL E A, CHOMISTEK A K, KINGMA J J, et al.Balance- and Strength-Training Protocols to Improve Chronic Ankle Instability Deficits, Part I: Assessing Clinical Outcome Measures[J]. J Athl Train,2018,53(6):568-577.
WISTHOFF B A, DOCHERTY C L, GLUTTING J, et al.Identifying Range-of-Motion Deficits and Talocrural Joint Laxity After an Acute Lateral Ankle Sprain[J]. J Athl Train,2021,56(4):408-417.
ALMEIDA M O, DAVIS I S, LOPES A D.Biomechanical Differences of Foot-Strike Patterns During Running: A Systematic Review With Meta-analysis[J]. J Orthop Sports Phys Ther,2015,45(10):738-755.
MCCANN R S, BOLDING B A, TERADA M, et al.Isometric Hip Strength and Dynamic Stability of Individuals With Chronic Ankle Instability[J]. J Athl Train,2018,53(7):672-678.
BROWN C N, BOWSER B, ORELLANA A.Dynamic postural stability in females with chronic ankle instability[J]. Med Sci Sports Exerc,2010,42(12):2258-2263.
WIKSTROM E A, TILLMAN M D, CHMIELEWSKI T L, et al.Dynamic postural control but not mechanical stability differs among those with and without chronic ankle instability[J]. Scand J Med Sci Sports,2010,20(1):e137-144.
ZHANG Z, ZHANG M.Effect of different ankle braces on lower extremity kinematics and kinetics following special-induced fatigue for volleyball players with functional ankle instability[J]. Heliyon,2023,9(6):e16380.
DELAHUNT E, MONAGHAN K, CAULFIELD B.Changes in lower limb kinematics, kinetics, and muscle activity in subjects with functional instability of the ankle joint during a single leg drop jump[J]. J Orthop Res,2006,24(10):1991-2000.
KIM H, SON S J, SEELEY M K, et al.Altered Movement Biomechanics in Chronic Ankle Instability, Coper, and Control Groups: Energy Absorption and Distribution Implications[J]. J Athl Train,2019,54(6):708-717.
FONG D T, CHAN Y Y, MOK K M, et al.Understanding acute ankle ligamentous sprain injury in sports[J]. Sports Med Arthrosc Rehabil Ther Technol,2009,1:14.

No competing interests reported.

Download PDF

Journal Publication

published 27 Sep, 2024

Read the published version in Scientific Reports →

Editorial decision: Revision requested
06 Aug, 2024
Reviews received at journal
05 Aug, 2024
Reviews received at journal
30 Jul, 2024
Reviewers agreed at journal
26 Jul, 2024
Reviewers agreed at journal
25 Jul, 2024
Reviewers invited by journal
23 Jul, 2024
Editor assigned by journal
23 Jul, 2024
Editor invited by journal
14 May, 2024
Submission checks completed at journal
14 May, 2024
First submitted to journal
29 Apr, 2024

You are reading this latest preprint version

Construction and Validation of a Risk Prediction Nomogram Model for Different Sites of Ankle Pain in Patients with Chronic Ankle Instability Using the OpenSim Environment

Status:

Journal Publication

Version 1

Abstract

Figures

WHAT IS ALREADY KNOWN ON THIS TOPIC

INTRODUCTION

METHODS

Participants

Equity, diversity, and inclusion statement

Test Apparatus

Motion capture system

Three-dimensional force plate

Surface EMG testing system

Experimental Design and Testing Procedures

Data Extraction and Preparation

Kinematics

Kinetics

Muscle force

Statistics

RESULTS

DISCUSSION

Clinical Implications

Limitations

CONCLUSION

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1