Association of Ankle Dorsiflexion and Landing Forces in Jumping Athletes

Abstract

Background:

Dorsiflexion range of motion restriction has been associated with patellar tendinopathy, but the mechanisms of how dorsiflexion restriction could contribute to knee overload remain unknown.

Hypothesis:

Peak ankle dorsiflexion and ankle dorsiflexion excursion are negatively associated with peak vertical ground-reaction force (vGRF) and loading rate, and with peak patellar tendon force and loading rate, and positively associated with peak ankle plantar flexor moment.

Study Design:

Cross-sectional study.

Level of Evidence:

Level 4.

Methods:

Kinematic and kinetic data of 26 healthy recreational jumping athletes were measured during a single-leg drop vertical jump. Pearson’s correlation coefficients were calculated to establish the association between peak ankle dorsiflexion and ankle dorsiflexion excursion with peak vGRF and vGRF loading rate, with peak patellar tendon force and patellar tendon force loading rate, and with peak ankle plantar flexor moment.

Results:

Ankle dorsiflexion excursion negatively correlated with peak vGRF loading rate (r = −0.49; P = 0.011) and positively correlated with peak ankle flexor plantar moment (r = 0.52; P = 0.006). In addition, there was a positive correlation between peak ankle dorsiflexion and peak vGRF (r = 0.39; P = 0.05).

Conclusion:

Ankle kinematics are associated with vGRF loading rate, ankle flexor plantar moment and peak vGRF influencing knee loads, but no association was observed between ankle kinematics and patellar tendon loads.

Clinical Relevance:

These results suggest that increasing ankle dorsiflexion excursion may be an important strategy to reduce lower limb loads during landings but should not be viewed as the main factor for reducing patellar tendon force.

Patellar tendinopathy (PT) has a high prevalence in sports, especially in those involving repetitive jumps. ²⁵ It has been reported that PT affects 31.9% and 44.6% of elite volleyball and basketball athletes, respectively. ¹⁰ The symptoms resulting from PT can be devastating for the athletes’ careers, with 53% of athletes with PT quitting sports participation because of their knee pain. ⁹

Overload has been shown to be a risk factor for PT development.^3,22 It is hypothesized that the biomechanics of the lower limb distal to the knee during jump landings can contribute to patellar tendon overload. ²⁴ Ankle dorsiflexion range of motion (ROM) (described as passive or active measures of dorsiflexion movement in clinical evaluations, ie, lunge test) limitation has been associated with PT in previous studies.^1,11,16,17 In a cross-sectional study, Malliaras et al, ¹¹ analyzing several strength and flexibility variables in volleyball athletes, found that only ankle dorsiflexion restriction was associated with PT. Similarly, Scattone Silva et al ¹⁶ showed that basketball, volleyball, and handball athletes with PT had lower dorsiflexion ROM than healthy athletes. Finally, Backman and Danielson ¹ in a 1-year prospective study reported that lower dorsiflexion ROM was the main risk factor for PT. However, a recent systematic review ²¹ concluded there is no strong evidence that ankle dorsiflexion limitation is a risk factor for PT development.

Although the association between dorsiflexion ROM limitation and PT has been observed, the mechanism of how dorsiflexion restriction could contribute to PT remains unknown. It is believed that the ankle dorsiflexion movement and eccentric calf muscles contraction are important aspects to absorb the forces acting on the lower limbs during jump landing, and impairment in this mechanism may result in higher patellar tendon load. ¹¹ Scattone Silva et al ¹⁷ found that athletes with PT have smaller peak ankle dorsiflexion during a drop landing task when compared with healthy athletes. However, to the authors’ knowledge, there are no studies evaluating the association between ankle dorsiflexion excursion (described as ankle movement during functional tasks, in this case, jump landings) with the vertical ground-reaction force (vGRF), patellar tendon force magnitude, and patellar tendon loading rate during jump landings. Considering that dorsiflexion restriction has been shown to cause a stiff jump landing pattern, with less hip and knee flexion,^4,27 it is possible that ankle dorsiflexion excursion restriction may be associated with higher landing forces and loading rate, which could contribute to knee extensor mechanism overload.

Therefore, the aim of this study was to evaluate the association between peak ankle dorsiflexion and ankle dorsiflexion excursion during landing with peak vGRF, vGRF loading rate, peak patellar tendon force, patellar tendon force loading rate, and peak ankle plantar flexor moment during a single-leg drop vertical jump. Our hypothesis is that peak ankle dorsiflexion and ankle dorsiflexion excursion are negatively associated with peak vGRF and loading rate, and with peak patellar tendon force and loading rate, and positively associated with peak ankle plantar flexor moment.

Methods

Participants

Twenty-six recreational jumping athletes (handball, basketball, and volleyball players) of both sexes who practiced sports at least 3 times a week were recruited from local sports teams between the years 2018 and 2019. After signing a written informed consent form, approved by the University Ethics Committee, all participants were evaluated for the inclusion and exclusion criteria. Healthy recreational athletes aged 18 to 35 years were included. Participants were excluded if they suffered from any injury in the lower limbs or lumbar spine in the past 12 months; reported pain or discomfort that interfered with the evaluations; or had neurological or vestibular disorders that prevented participation in the study. Recruitment was performed through flyers in social media, universities, and direct contact with clubs.

A priori sample size calculations were conducted for correlation data using the G*Power software (Version 3.1.9.2). Considering a moderate-to-good correlation (r = 0.5; with α = 0.05 and β = 0.2), the calculations estimated that a minimum of 23 participants would be necessary for this study.

Procedures

Participants were instructed to avoid nonhabitual physical activity 48 hours before the evaluation. For the evaluation, participants wore a top (female participants), shorts, and neutral athletic shoes (Asics Gel-Equation 5), provided by the examiner. Only the dominant lower limb was assessed. The dominant lower limb was determined by asking the participants which leg they used to kick a ball. ¹⁷

Single-Leg Drop Vertical Jump

The biomechanical evaluation was performed using a 3-dimensional motion analysis system (Vicon Motion Systems Ltd) with 6 cameras synchronized with an AMTI Force and Motion force plate (Model OPT400600HF-2000) during a single-leg drop vertical jump task. For data acquisition, a sampling rate of 240 Hz for the kinematic data and of 1200 Hz for the kinetic data were used. The Nexus System 2.9.3 software (Vicon Motion Systems Ltd) and the Motion Monitor software (Innovative Sports Training) were used for data analysis. After the system calibration, the same researcher positioned reflective markers (14 mm diameter) on anatomic landmarks at iliac crest bilaterally, anterior superior iliac spine and posterior superior iliac spine bilaterally, first sacral. For dominant lower limb, landmarks were over greater trochanter, medial and lateral femoral condyles, medial and lateral malleoli, immediately over second and fifth metatarsal heads on the shoe, immediately over both calcaneus on the shoe, and fifth metatarsal base and 2 clusters on the lateral aspect of the thigh and leg of the participants. ¹² Then, the participants were positioned in the center of the force plate and a static calibration was performed to align the participant with the global coordinate system. For the single-leg drop vertical jump task, the participants were positioned over a 31-cm box crossing their arms over the chest and instructed to step off the box landing with the dominant lower limb in the center of the force plate. Immediately after touching the force plate, the participants were instructed to perform a maximal effort single-leg vertical jump. ¹² All participants performed 5 single-leg drop vertical jumps for familiarization purposes. After 2 minutes of rest, 5 valid trials were recorded. The single-leg drop vertical jump was considered valid if the participants performed the task with the arms across the chest, stepping off the box without jumping up, stepping down, or losing balance; and the landing occurred in the center of force plate. During the test, no visual or verbal cues or landing instructions were given to the participants.

Data Analysis

Kinematic and kinetics data processing were performed with the Motion Monitor software, which was used for the creation of the biomechanical model of the body segments. Euler angles were calculated using the joint coordinate system recommendations from the International Society of Biomechanics.^6,26 Knee joint center was considered as the midpoint between the medial and lateral femoral epicondyles and ankle joint center as the midpoint of the medial and lateral malleoli. Knee flexion (used to calculate peak patellar tendon force) was evaluated as the angle between shank and thigh and ankle dorsiflexion was evaluated as the angle between foot and shank, both in the sagittal plane. Kinematic and kinetic data were filtered using a second-order zero-lag Butterworth 12-Hz low-pass filter and fourth-order zero-lag Butterworth 50-Hz low-pass filter, respectively.

Peak knee extensor and ankle plantar flexor moments (internal joint moments) were calculated by inverse dynamics. At this stage, MatLab version 9.0.0.341360, R2016a (Mathworks) software was used for data processing of the variables of interest. Variables of interest were extracted during the landing phase, which was defined as the period between the initial contact of foot with force plate—when vGRF first exceeded 10 N—and peak knee flexion. Variables of interest were peak ankle dorsiflexion, ankle dorsiflexion excursion (defined as the difference between peak ankle dorsiflexion and the ankle angle in the initial foot contact), peak ankle flexor plantar moment normalized by body weight, peak vGRF normalized by body weight; vGRF loading rate, peak patellar tendon force normalized by body weight, and patellar tendon force loading rate. vGRF loading rate was defined as normalized peak vGRF divided by the time to reach this peak. Peak patellar tendon force was calculated as the knee joint moment divided by the patellar tendon moment arm, ¹⁴ estimated by a regression equation using knee flexion angle. ⁷ Patellar tendon loading rate was defined as normalized peak patellar tendon force divided by the time to reach this peak. ⁸ The average of the 5 repetitions was used in the analyses.

Statistical Analysis

All statistical analyses were conducted using Statistical Package for the Social Sciences (SPSS Version 19.0.0, IBM Corp). Initially, the statistical distribution and homoscedasticity were verified with the Shapiro-Wilk and Levene tests, respectively. A Pearson correlation matrix was used to investigate the association between each dependent variable (peak ankle flexor plantar moment, peak vGRF, vGRF loading rate, peak patellar tendon force, and patellar tendon force loading rate) and the independent variables (peak ankle dorsiflexion and ankle dorsiflexion excursion). The significance level was set as 0.05.

Results

Descriptive results regarding the demographic data are presented in Table 1. The results of kinematics and kinetics during the single-leg vertical jump are shown in Table 2.

Table 1.

Demographic characteristics of the subjects (mean ± SD), n = 26

Demographic data
Male/female, n	21/5
Age, y	23.62 ± 4.24
Height, m	1.76 ± 0.08
Body mass, kg	77.42 ± 13.28
Sports participation frequency (times/week)	4.11 ± 1.03
Sports (number of athletes)
Handball	8
Basketball	7
Volleyball	11

Table 2.

Biomechanical data of subjects expressed in mean ± SD (range)

Joint angle, deg
Peak ankle dorsiflexion	16.50 ± 4.30 (7.89-26.06)
Ankle dorsiflexion excursion	46.22 ± 8.65 (29.17-58.77)
Joint moment, N·m/BW
Peak ankle plantar flexion moment	0.18 ± 0.04 (0.11-0.29)
Forces
Peak vGRF, N/BW	1.93 ± 0.32 (1.14-2.41)
Peak patellar tendon force, BW	3.94 ± 1.23 (2.12-7.32)
vGRF loading rate, BW/s	30.37 ± 9.65 (13.32-38.57)
Patellar tendon loading rate, BW/s	25.80 ± 10.74 (10.10-46.05)

BW, body weight; vGRF, vertical ground-reaction force.

Ankle dorsiflexion excursion positively correlated with peak ankle flexor plantar moment (r = 0.52; P = 0.006) and negatively correlated with vGRF loading rate (r = −0.49; P = 0.011). However, there was no correlation between ankle dorsiflexion excursion and peak vGRF, peak patellar tendon force or patellar tendon force loading rate (Table 3). A positive correlation between peak ankle dorsiflexion and peak vGRF was also observed (r = 0.39; P = 0.05). Finally, there were no correlations between peak ankle dorsiflexion and peak ankle plantar flexor moment, vGRF loading rate, peak patellar tendon force, or patellar tendon force loading rate (P > 0.05) (Table 3). Diagrammatic representations of the significant correlation are presented in Figure 1.

Table 3.

Pearson correlation coefficients (r) between peak ankle dorsiflexion, ankle dorsiflexion excursion, and the kinetic variables

	Peak Ankle Dorsiflexion	Ankle Dorsiflexion Excursion
Peak ankle plantar flexor moment
r	0.332	0.52
P	0.10	0.006*
Peak vGRF
r	0.39	0.276
P	0.05*	0.17
vGRF loading rate
r	0.032	−0.489
P	0.88	0.011*
Patellar tendon loading rate
r	−0.004	0.188
P	0.99	0.36
Peak patellar tendon force
r	0.039	0.330
P	0.85	0.10

vGRF, vertical ground-reaction force.

^*Statistically significant (P > 0.05).

Figure 1.

Diagrammatic representation of the significant association between peak ankle dorsiflexion (A), ankle dorsiflexion excursion (B-C) and kinetics variables. BW, body weight; vGRF, vertical ground reaction force.

Discussion

A previous study ¹ has observed that limited ankle dorsiflexion ROM is a risk factor for PT, and it has been hypothesized that this movement restriction would result in increased patellar tendon load during tasks such as jump landings. ¹¹ The main result of this study was the fact that a lower ankle dorsiflexion excursion during landing was associated with greater vGRF loading rate. Positive associations were also observed between ankle dorsiflexion excursion and peak ankle plantar flexor moment and between peak ankle dorsiflexion with peak vGRF. However, there was no association between ankle dorsiflexion excursion and peak ankle dorsiflexion and the other kinetic variables.

In the current study, ankle dorsiflexion excursion during landing had a moderate correlation with vGRF loading rate, which indicates that ankle movements have a significant impact on the rate at which force is dissipated. To our knowledge, no other study has investigated the association between ankle kinematics during landing and vGRF loading rate, which makes comparison of our results difficult. However, when landing from a single-leg drop vertical jump (toe-to-heel landing pattern), the ground-reaction force creates an external dorsiflexor moment, and to prevent excessive dorsiflexion, the plantar flexors muscles generate a plantar flexor internal moment. ²⁰ Previous studies^4,27 have confirmed that the ankle joint makes a substantial contribution to ground-reaction force dissipation. Devita and Skelly ⁴ concluded that the ankle plantar flexors muscles are responsible for the majority of energy absorption during jump landings, with an average of 44% of the total muscular work. Thus, restricted ankle dorsiflexion excursion may limit the contribution of the plantar flexor muscles to exert force during landing, with the ankle potentially becoming less efficient near end of range. ¹¹ This may result in lower limb overload and potentially increases the risk of injury. ¹³ The positive association between ankle dorsiflexion excursion and ankle plantar flexor moment, observed in the current study, supports the hypothesis that ankle dorsiflexion excursion influences the ability of the ankle plantar flexors to dissipate the landing forces. In this context, strategies to increase ankle dorsiflexion excursion (such as calf muscles stretching and posterior talus mobilizations) and the eccentric plantar flexors strength could be used for reducing the vGRF loading rate.

Ankle dorsiflexion excursion was not correlated with peak vGRF in the jumping athletes of the current study. However, the negative association between ankle dorsiflexion excursion with vGRF loading rate may be a more relevant result regarding overload injuries, since vGRF loading rate is a derivation of vGRF in time during load absorption while the peak vGRF is just raw value of load. Because human tissue is viscoelastic, its loading response is time dependent, and the tissues are more prone to injury when submitted to higher loading rates. ⁵ In this sense, the human tissues are more prone to injury when submitted to high loading rates. In this context, we hypothesize that jump landings with greater ankle dorsiflexion excursion may be beneficial to reduce the vGRF loading rate and potentially help prevent overuse injuries. However, prospective studies are necessary to confirm this hypothesis.

Interestingly, although greater ankle dorsiflexion excursion correlated with lower vGRF loading rate, there was no association between ankle kinematics and peak patellar tendon force and patellar tendon loading rate. Therefore, the hypothesis that greater ankle dorsiflexion excursion would decrease patellar tendon load was not confirmed. Janssen et al ⁸ found that individual factors such as ankle kinematics alone were not associated with peak patellar tendon force during jump landings. However, the authors observed that greater ankle dorsiflexion velocity, male sex, greater quadriceps strength, and greater trunk flexion velocity during landing were factors that, combined, were able to predict patellar tendon force and loading rate. ⁸ However, ankle dorsiflexion excursion was not a variable in this study. Factors proximal to the knee have been shown to influence knee loads.^2,17,19 For example, Shimokochi et al ¹⁹ found that increasing trunk and hip flexion during single-leg jump landings decreases the knee extensor moment and vGRF in healthy subjects. Similarly, Blackburn and Padua ² have shown that greater trunk flexion during a double-leg landing reduced the vGRF in healthy subjects. More recently, Scattone Silva et al ¹⁷ observed that a single set of verbal instructions to land with more trunk flexion resulted in lower peak knee extensor moment and peak patellar tendon force in elite volleyball and basketball athletes with and without PT. In this context, the subjects of the present study that had lower ankle dorsiflexion may have used a strategy of increasing trunk and hip flexion to dissipate the vGRF to minimize the loads on the patellar tendon. This could explain the lack of association between ankle dorsiflexion excursion with the peak patellar tendon force and the patellar tendon load rate.

Against to our initial hypothesis, we found a positive association between peak ankle dorsiflexion and peak vGRF. It is possible that peak ankle dorsiflexion alone is not a relevant variable considering lower limb load absorption during a jump landing. Rowley and Richards ¹⁵ observing landings with different initial contact ankle plantar flexion angles have shown that increased plantar flexion during initial contact was associated with decreased peak vGRF and vGRF loading rate. This can be explained by the greater amount of energy absorbed by eccentric contraction of the gastrocnemius and soleus muscles. ¹⁸ Thus, it was possible that, although some subjects in our study presented higher peak ankle dorsiflexion during landing, the initial contact may have occurred with less ankle plantar flexion, which led to less vGRF absorption. Considering this possibility, a secondary analysis was conducted, and we observed that lower ankle plantar flexion angle at initial contact was associated with lower ankle plantar flexor moment (r = 0.41, P < 0.04) and greater vGRF loading rate (r = −0.59, P < 0.01), which supports this hypothesis.

Significant limitations should be acknowledged when interpreting our findings. Passive ankle ROM was not assessed. Considering that this is a low-cost assessment procedure that can easily be performed in clinical practice, future studies should evaluate whether a relationship exists between passive ankle dorsiflexion ROM and peak patellar tendon force and loading rate. Only healthy recreational athletes were evaluated and the extrapolation of these results to other population should not be done. Considering the existence of sex differences in lower limb biomechanics, the inclusion of male and female participants could be considered a limitation of this study. However, we decided to include both male and female athletes to increase the study’s external validity. Also, in a secondary analysis, excluding the female participants, the results did not change. In addition, a 2-dimensional model of patellar tendon force calculation is very limited and may overestimate the real patellar tendon force since the patellar tendon has a tridimensional structure. Only ankle kinematics was considered in this study and others joint movements may better explain patellar tendon load. Another important study limitation was that the use of a biomechanical model based on the inverse dynamics did not allow the analysis of the coactivation of the muscles around the knee, ²³ which may significantly underestimate the forces in the patellar tendon. Last, our study design indicates association between variables, but it cannot determine causality.

Conclusion

Greater ankle dorsiflexion excursion was associated with smaller vGRF loading rate in jumping athletes. In addition, a positive association between ankle dorsiflexion excursion and peak ankle plantar flexor moment was observed during a single-leg drop vertical jump task, with no association between ankle kinematics and patellar tendon forces. These results suggest that increasing ankle dorsiflexion excursion may be an important strategy to reduce lower limb loads during landings but should not be viewed as the main factor for reducing patellar tendon loads.