Physical Therapy Korea

Ankle sprain is a common musculoskeletal injuries especially in young females with disabling symptoms, such as pain, tenderness, swelling, gait dysfunction, and instability of the ankle joint [1]. Ankle sprains occur with an incidence rate of 2.15 to 3.29 per 1,000 person each year among the general population in the United States [1-3]. A meta-analysis on the prevalence of ankle sprain concluded a higher incidence of ankle sprain in females compared with males (13.6 vs 6.94 per 1,000 exposures) [4]. A systematic review of 31 follow-up studies reported that 5% to 33% of ankle injury patients still experienced pain after one year; and 33% of the patients reported at least one re-sprain within a 3-year period [5].

Lateral ankle sprain is more prevalent and spends greater medical expenses than medial ankle sprain [1]. Lateral ankle sprain occurs with greater plantar flexion at the talocrural joint and foot inversion at the subtalar joint and internal rotation at the transverse tarsal joint [6-8]. These motions are the composites of foot supination. The primary ligamentous restraint to an inversion moment in a plantar flexed position is the anterior talofibular ligament [9]. Anticipatory responses to induced inversion perturbations pose a considerable challenge in a laboratory research, and recent evidence suggests that anticipating inversion perturbations result in significant alterations to lower extremity movement dynamics [10].

Sports with frequent jumping and landing such as basketball, volleyball, and soccer have been reported to have the highest rates of lateral ankle sprain injury [5]. Landing is a basic motor skill, and it is needed every day for walking, stair negotiation, jumping and running. The effective and efficient performance of these movements found to be matured at about 10–12 years of age [11]. The objective of landing is to absorb the kinetic energy of the body, while maintaining balance with the spatiotemporal prediction of the ground contact, the prediction of the magnitude of ground reaction force (GRF), and the control of the position and angular displacement of multiple joints by activating or inhibiting the appropriate muscles in the right timing and magnitude [12]. Pre-activation of anti-gravity muscles plays a major role on the landing strategy: anticipatory joint positioning and joint stiffness [12]. The pre-activation mechanism occurs with information about the time to contact (τ, Tau) with a surface in landing and vision is a major source of information [13]. The magnitude of impact is primarily a function of the height of the fall, which determines the velocity of impact and the severity of injury [14]. With higher landing height, the amplitude of muscle pre-activation increases [15-20] but the duration of those pre-activation is not affected [12,21]. More importantly, the joint stiffness and the resultant GRF are affected by whether the landing is with or without precise visual feedback [18,20]. However, the landing heights in some previous studies [13,17,18,22,23] were way beyond (up to 2 m) the usual height of daily activities (most of ankle inversion sprain does not occur in landing from very high steps!) and the participant had enough time to prepare themselves for the forthcoming landing. Once the participants are aware of the height of landing and there is sufficient time to prepare [17,24], they can activate their long-mastered landing motor program and perform smooth landing even without the ongoing visual feedback [13]. With enough time for preparation, we don’t sprain our ankles when landing from the considerably high steps without looking at it. When an incorrect estimation and preparation of landing force and timing might lead to serious injuries the joints of landing limb [14,25,26].

Anticipatory postural adjustment of the landing lower limb is a key modifiable factor in preventing the lateral ankle sprain. Changes in limb segment orientation relative to the landing surface change the alignment of the GRF vector relative to the joints, hence the joint moments [12]. Increased ankle plantar flexion was found when landing onto hard rather than soft surfaces, resulting in larger ankle joint excursion after the touch down [27]. Step height was not significantly affect the kinematics of lower limb at the touch down, especially when the height was less than 40 cm [18]. Removal of visual feedback did not have significant effect on either pre-landing electromyography (EMG) amplitude or lower limb kinematics at the touch down. However, it is not clear whether the removal of visual feedback changes the movement of landing lower limb in the air when the landing was initiated voluntarily nut not with enough time to plan the movement.

In this study, the tested landing heights were not higher than the usual height of daily activities (20 cm, a height for standard stairs, and 40 cm, a bus step height) with and without blindfolding. In addition, the landing foot was randomly announced as a starting signal and the landing on that side was initiated immediately to minimize the time to plan their movement. This study aimed to investigate the effect of landing height and the vision on the kinematics of landing and supporting lower limbs and the GRFs. Findings of this study can contribute to understanding of the injury mechanisms and preventing the lateral ankle sprain.

1. Participants

Eighteen healthy females without significant history of injury or surgery in the lower extremities were voluntarily participated in this study. Participants with prior experience of major ankle sprain and neurological disorder were excluded. The females were found to have different landing mechanics [19,28,29] and had more frequent lateral ankle sprains [2,4]. It was found that females and males differ in their anticipatory postural control strategy [30]. They were 22.15 ± 2.65 years, 162.15 ± 5.60 cm, and 57.75 ± 8.32 kg. All who volunteered to participate in this study signed a consent form describing the procedure and the purpose of this study before the commencement of the experiments. This study was approved by the Institutional Review Board of Hoseo University (IRB no. 1041231-210504-HR-124).

2. Testing Procedure

After explaining the testing procedure and signing the informed consent, all participants practiced the landing from the lower (20 cm) and the higher (40 cm) steps several times with and without visual feedback for familiarization with the tasks prior to data collection. Standard indoor step height is between 7 and 7 ¾ inches (17.78 and 19.69 cm) in US [31] and the maximum height of 40 cm is for the first step of the public bus [32]. These step heights were chosen to figure out the double dose effect of landing height. The arms were crossed in front of the chest and the trunk was kept in straight position to avoid any excessive arm motion. Participants stood up on the one of two steps with bare feet and an investigator was within an arm reach range for safety. Their eyes were blindfolded in no visual feedback condition, and they were asked to look down in visual feedback condition. They were instructed to step down softly with the left or right foot on the 40 × 60 cm force plate, 2 cm away from the front edge of step. Stepping down foot was randomly assigned to the participants when they are ready, and they were asked to initiate the movement immediately. If they were not able to start stepping down immediately after announcement, the trial was repeated. All participants completed 24 trials (2 landing heights; 2 visual conditions; 2 stepping sides; and 3 repetitions per condition) in a randomized order.

3. Data Collection

A multi-component force platform (MBTI, Kistler; resonant frequency in situ: 1,000 Hz) was used for measuring the GRF. The force platform signal allowed the measurement of peak vertical and its timing. The force platform signal was normalized to the body weight of participants. Polhemus Liberty (Polhemus) motion tracking system with eight active electromagnetic sensors was used to collect the kinematic data of bilateral lower limbs at 240 Hz. The local coordinate systems recommended by the International Society of Biomechanics [33] were used to describe joint motions. The sensors were securely attached to the sacrum, thighs, shanks, and dorsum of the feet with double-sided tape and Velcro straps (VELCRO®) and each segmental axis was configured based on the ISB protocols. The MotionMonitor integrated motion capture software (Innovative Sport, Inc.) calibrated the kinematic and GRF data in accordance with the company guidelines.

Event points of landing in this study were 5 (Figure 1): Point 1 (P1) was defined when the landing heel beginning to take off from the step; Point 2 (P2) was when the landing toe-off the step and the supporting knee beginning to flex; Point 3 (P3) was when the landing toe touching down on the force plate; Point 4 (P4) was when the landing heel touching down and the supporting knee flexing maximally; and Point 5 (P5) was when the landing knee beginning to extend and the supporting toe-off the step. All the joint angles were zeroed out at P1.

Figure 1. Event points (1–5) of the right-side landing in this study.

4. Statistical Analysis

The data were analyzed using two-way ANOVA with repeated measures to determine whether the step height, the visual feedback, or the interaction between these two factors had a significant effect on the landing duration, the kinematic variables and the GRFs at each event point (IBM SPSS 19.0, IBM Co.). The kinematics of landing and supporting knee, ankle and foot were analyzed separately. The significance level was set at 0.05. The distribution of all dependent variables was examined by using the Shapiro–Wilk test and was found not to differ significantly from normality.

1. Landing Duration

Landing duration in this study was defined as the times spent from P1 to P5, from heel-off of the landing foot to toe-off of the supporting foot from the step. The overall landing duration was 1.50 ± 0.64 seconds (Figure 2). Landing from the higher step (1.60 ± 0.68 seconds) took longer (main effect of landing height, F = 10.84, p = 0.01) than landing from the lower step (1.40 ± 0.58 seconds). The blindfolded vision had no significant main effect (F = 0.14, p = 0.71) on the landing duration with insignificant (F = 0.88, p = 0.35) interaction between step height and visual feedback. The proportional time while the landing foot is in the air, % time between P1 and P2 over the landing duration, were increased from the lower step (F = 14.49, p < 0.01) and with blindfold (but not statistically significantly, F = 3.32, p = 0.06). No other % time were affected by landing height nor visual feedback.

Figure 2. Landing duration (A) and cumulative proportional time (B) at each landing point. Lighter line is landing from the lower step and darker line is landing from the higher step. Dotted line is landing with eyes open and solid line is landing with eyes blindfolded. *Significant (p < 0.05) effect of the landing height at that point.

2. Knee Flexion

Landing knee flexion had double peaks: at P2 and P4, while supporting knee flexes maximally at P4. With eyes blindfolded, the landing knee flexed more at P2 (F = 4.58, p = 0.04), P3 (F = 4.39, p =0.04), and P5 (F = 9.75, p < 0.01). Landing knee flexions at P2, P3, P4, and P5 were 46.13° ± 12.02°, 3.12° ± 9.43°, 26.09° ± 12.76°, and 1.79° ± 9.76° with eyes open, while 48.42° ± 10.07°, 5.03° ± 9.71°, 27.80° ± 12.99°, and 4.62° ± 9.00° with eyes blindfolded (Figure 3). From the higher step, the landing knee flexed more at P4 (F = 106.95, p < 0.01). Landing knee flexions at P2, P3, P4, and P5 were 47.52° ± 11.25°, 4.67° ± 10.30°, 21.20° ± 12.75°, and 2.76° ± 9.73° from the lower step, while 48.42° ± 11.03°, 5.03° ± 8.84°, 27.80° ± 10.21°, and 4.62° ± 9.23° from the higher step. There was no significant interaction between the landing height and the visual feedback on the landing knee flexion.

Figure 3. Knee flexion of landing (A) and supporting (B) limbs at each landing point. Lighter line is landing from the lower step and darker line is landing from the higher step. Dotted line is landing with eyes open and solid line is landing with eyes blindfolded. *Significant (p < 0.05) effect of the landing height and **significant effect of the visual feedback at that point.

Supporting knee progressively flexed until P4 and quickly extended. Higher landing height increased the supporting knee flexion at P2 (F = 56.05, p < 0.01), P3 (F = 391.75, p < 0.01) and P4 (F = 452.31, p < 0.01). There was no significant main effect of the blindfold or interaction between the landing height and the visual feedback on the supporting knee kinematics.

3. Ankle Dorsiflexion

Landing ankle dorsiflexion had double peaks: firstly, at P2 and secondly, at P4. With eyes blindfolded, the landing ankle dorsiflexion decreased (more plantar flexed, F = 4.36, p = 0.03) to 1.39° ± 5.90° from 0.18° ± 6.01° with eyes open at P2 but increased to –3.17° ± 5.08° from –5.17° ± 5.28° (F = 17.53, p < 0.01) at P5. Landing ankle plantar flexion was maximal at P3. Higher landing height increased (F = 4.55, p = 0.03) the landing ankle plantar flexion from 42.05° ± 14.06° to 44.86° ± 12.95° at P3 (Figure 4). Higher landing height increased dorsiflexion at P4 (F = 105.85, p < 0.01) and P5 (F = 42.74, p < 0.01). There was no significant interaction between the landing height and the visual feedback on the landing ankle dorsiflexion.

Figure 4. Ankle dorsiflexion of landing (A) and supporting (B) limbs at each landing point. Lighter line is landing from the lower step and darker line is landing from the higher step. Dotted line is landing with eyes open and solid line is landing with eyes blindfolded. *Significant (p < 0.05) effect of the landing height and **significant effect of the visual feedback at that point.

Supporting ankle progressively dorsiflexed until P3 and it plantar flexed. Higher landing height increased the supporting ankle dorsiflexion at P2 (F = 46.51, p < 0.01), P3 (F = 5.00, p = 0.03) and P4 (F = 5.70, p = 0.02). There was no significant main effect of the blindfold or interaction between the landing height and the blindfold on the supporting ankle kinematics.

4. Foot Inversion

Landing foot inverted at P2 and P3. Landing from the higher step increased (F = 4.31, p = 0.04) the foot eversion at P4. There was no significant main effect of the blindfold or interaction between the landing height and the blindfold on the landing foot inversion. Landing foot inversions at P2, P3, P4, and P5 were 12.46° ± 6.95°, 13.37° ± 9.53°, 1.37° ± 8.69°, and 1.28° ± 7.88° from the lower step, while 12.28° ± 5.50°, 12.57° ± 9.63°, –0.45° ± 9.42°, and 0.14° ± 8.70° from the higher step (Figure 5). Landing foot inversions at P2, P3, P4, and P5 were 12.33° ± 6.17°, 12.64° ± 9.72°, –0.24° ± 8.85°, and 0.38° ± 8.46° with eyes open, while 12.61° ± 6.37°, 13.30° ± 9.44°, 1.16° ± 9.31°, and 1.04° ± 8.16° with eyes blindfolded.

Figure 5. Foot inversion of landing (A) and supporting (B) limbs at each landing point. Lighter line is landing from the lower step and darker line is landing from the higher step. Dotted line is landing with eyes open and solid line is landing with eyes blindfolded. *Significant (p < 0.05) effect of the landing height at that point.

Supporting foot increasingly inverted from P2 and P4. There was no significant effect of the landing height or blindfold on the supporting foot kinematics.

5. Normalized Ground Reaction Force

Landing from the higher step increased the vertical normalized ground reaction force (NGRF) at P5 (F = 172.12, p < 0.01), the forward NGRF at P3 (F = 23.07, p < 0.01), and the lateral NGRF at P4 (F = 5.31, p = 0.02) and P5 (F = 11.87, p < 0.01). There was no significant main effect of the blindfold or interaction between the landing height and the blindfold on the NGRF at any point of landing (Figure 6).

Figure 6. GRFs (%BW) on the vertical (A), anteroposterior (B), and mediolateral (C) axes at each landing point. Lighter line is landing from the lower step and darker line is landing from the higher step. Dotted line is landing with eyes open and solid line is landing with eyes blindfolded. GRF, ground reaction force; BW, body weight. *Significant (p < 0.05) effect of the landing height at that point.

This study aimed to investigate the effect of landing height and the vision on the kinematics of landing and supporting lower limbs and the GRFs. The landing heights were a usual stair height of 20 cm and a bus step of 40 cm. In addition, the landing foot was randomly assigned to minimize the time for planning the landing movement. With eyes blindfolded, the knee flexion and ankle dorsiflexion on landing side significantly decreased before and after the touch down. However, there was no significant effect of landing height on the anticipatory kinematics on the landing side. The step height had effect mainly on the supporting side and the landing side after the touch down. From the higher steps, the GRF increased but not without visual feedback. No interaction between step height and visual feedback was significant for all dependent variables in this study.

1. Landing Height

Prior to the touch down, the landings from the higher step increased the knee flexion and ankle dorsi flexion on the supporting side. However, there was no significant effect of landing height on the anticipatory kinematics on the landing side. Landing from the higher step increased the followings: landing duration; landing knee flexion at P4; supporting knee flexion at P2, P3, and P4; landing knee flexion at P4; supporting knee flexion at P2, P3, and P4; landing ankle plantar flexion at P3 and dorsiflexion at P4 and P5; supporting ankle dorsi flexion at P2, P3, and P4; normalized vertical GRF at P5; normalized anterior GRF at P3; and normalized lateral GRF at P4 and P5. Landing from the lower step increased the proportional time between P1 and P2.

Obviously, the higher landing height has a major influence on the landing duration, the kinematics of landing and supporting limbs (more extensively on the supporting side), and the GRF as in the previous studies [16-18,20]. When stepping down from a certain height, a person precisely estimates the impact force and its timing of landing from his or her successful motor learning experience and prepares by pre-activating the antigravity muscles and by modifying the joint angles to control the joint stiffness of landing limb [16,18-20,34,35]. Theoretically, adjusting the landing limb segments relative to the direction of the resultant GRF can scale the magnitude of joint reaction forces encountered at the touch down [12]. Changes in landing limb segment orientation in the air can modulate the alignment of the GRF vector relative to the joints, to decrease the joint moments and the risk of injury [17,36,37]. However, the landing height in this study did not affect the landing knee extension or ankle plantar flexion at P2 while the landing foot is in the air, but these key component in controlling the impact force of landing were affected when the eyes were blindfolded. Without vision, the landing knee was less flexed and the ankle was less dorsiflexed at P2, and the knee was more extended at P3. It means the participants did not move the landing knee and ankle same way without ongoing visual feedback. The landing height in this study might not high enough to change the limb orientation in the air.

Landing height more extensively affected the supporting limb and the GRF than the visual feedback. Kinematics of the supporting limb has not been a main concern in many landing studies. In case of drop landing [10,12,16,17,22,23,26,36], there are no supporting limb. Duncan and McDonagh [38,39] found that the post-landing muscle activity is a ‘programmed’ response rather than a reflex response to stretch of anti-gravity muscles. Gymnasts exhibited more knee flexion before and at ground contact, but less knee flexion at maximum knee flexion position resulting in higher vertical GRF and shorter braking phase during landing [17]. To control the landing impact and risk of injury, eccentric contraction of the supporting limb as well as preparatory contraction of the landing limb should cooperate with each other. If the anti-gravity muscles on the supporting limb are not strong enough, the impact force and the risk of injury could be greater, especially in the elderly. There should be more effort in studying the biomechanics of the supporting limb with various age group in the future.

2. Visual Feedback

Knee flexion and ankle dorsiflexion on landing side significantly decreased without visual feedback before and after the touch down in this study. With eyes blindfolded, the followings were significantly decreased in this study: landing knee flexion at P2, P3, and P5 and the landing ankle dorsiflexion at P2 and P5. The participants in this study were skillful enough to estimate the impact and to prepare themselves for those step height (20 and 40 cm). Without ongoing visual feedback, they slightly hesitated and mitigated their landing limb movement when it is in the air. Once the landing limb touches down the ground, visual feedback does not affect the kinematics of landing and supporting limbs or GRF anymore, while the effect landing height became much stronger after the touch down.

In drop landings from 10–130 cm steps, the ongoing visual feedback decreased the variability of pre-landing muscle activity [40], but it does not play a major role in triggering the preparatory actions in the self-initiated falls [13]. Once the participants already awarded the step height and a structured landing plan has been acquired, the authors concluded that the relevant muscles respond relative to the initial condition of the fall. A kinematic study with a moving platform and blindfold for the participants to unaware of the actual landing height [18] found that little or no modulation of lower limb joint angles occurs during landing from 20 to 80 cm step height and suggest that different control strategies were used to compensate for the lack of ongoing visual feedback. Without vision, the ongoing sources of sensory information available to control the movement during the fall are of proprioceptive and vestibular origin but these are not as sensitive or accurate as vision. Once a person already knows the step height prior to the landing and has a ‘programmed’ motor skill from the previous experience, a smooth landing from a 20 cm or a 40 cm step is not very difficult even when the eyes were blindfolded. The participants in this study practiced the landing from these step heights several times with and without visual feedback through the familiarization session prior to data collection. However, they might not have enough time to plan for a smooth landing in this study since they did not know which side to land until the last moment. The decreased knee flexion at P2 and P3 with the eyes blindfolded can be an evidence of ineffective motor strategy of stiff landing as in previous studies [17,41]. Ongoing visual feedback was required to regulate pre-landing EMG activity [24,42] and postural control in the air [43,44].

3. Anticipatory Postural Adjustment

Anticipatory postural adjustment is a high-level motor skill acquired through a motor learning process. It is specific to a task and does not generalize across tasks [45]. The dynamics of an expected perturbation modulate the kinematics and kinetics prior to an impaction [15]. Changes in landing limb orientation in the air relative to the landing surface change the alignment of the GRF vector relative to the joints, hence the joint moments [12,46,47]. Specifically, the greater ankle plantar flexion results in larger ankle joint excursion to absorb the landing impact effectively after foot contact [16,18,20]. When ankle plantar flexion is too much at the time of the touch down; however, the foot tends to invert and the risk of inversion sprain increases [48]. Sufficient ankle plantar flexion without precarious foot inversion at the time of initial contact is important in safe and smooth landing. Simpson et al. [10] anticipated inversion perturbations alters the ankle joint kinematics and impact kinetics during a single-leg drop landing. Visual feedback seems to be a key element for accurate estimation and execution of safe stepping down the stairs, especially for the first couple of steps where the proprioceptive memory is not firmly established. From my personal clinical experience, many patients with ankle inversion sprain stated that it happened on the first or second step of the stairs.

All the interaction effect between landing height and visual feedback were insignificant in this study. It means the effects of landing height and visual feedback are ‘independent’ in the landing tasks of this study. Stepping down from a 20 cm and a 40 cm step was a nothing new task for the participants in this study. That is why the landing duration was only increased by landing height but not by vision. Christoforidou et al. [17] found the absolute duration between P3 and P4 (in their study, defined as ‘braking phase’) and the pre-activation duration increased significantly for all muscles they tested with the drop height greater than 60 cm. The impact forces of landing were affected only by landing height but not by visual feedback [40]. No interaction effect between visual feedback and landing height was significant in those studies.

4. Limitation of Study

Only healthy young female participants were recruited due to their high prevalence of ankle inversion sprain, especially during descending stairs. In the male, ankle sprain occurs more in sports activities, such as basketball, football and soccer [1,2]. During drop landing from higher height, there was no sex difference in the landing strategy and the joint energetics [37]. Analyses of inverse dynamics, in which the direct and indirect joint loading can be estimated as a joint moment, are not included in this study. Muscle activity during the preparatory and responsive phase of the landing was not collected in this study. In the majority of previous landing studies [12,13,16-18,20,22,23,28,35,36,40,49,50], it is a major outcome presented. However, in focusing on the risk of ankle inversion sprain, current study analyzed the kinematics of landing and supporting limbs and the GRFs without monitoring the muscles around the ankle.

Step height and visual feedback affected the lower limb kinematics independently. Visual feedback affected on the landing lower limb prior to the touch down while step height had effect on the supporting side. After the touch down, the step height had more extensive influence on the lower limb kinematics and the GRFs than the visual feedback. The findings from this study could provide valuable insights into specific aspects of the risk underlying lateral ankle sprains.

Special thanks to Dami Yang, Jihae Yoo, Sohyun Chio, and Taehong Kim (R.I.P.) for their support in data collection and analysis.

None to declare.

No potential conflict of interest relevant to this article is reported.