Subjects
Twenty elite male paratroopers (mean age, 22.56±3.76 yr; mean height, 174.32±4.58 cm; mean weight, 62.42±6.93 kg) with formal parachute landing training and over 2 yr of parachute jumping experience volunteered for this study. All eligible subjects were healthy and had no history of lower extremity trauma or spinal fractures. None of the subjects had a history of previous surgery of the lower extremities, neurological or joint degenerative diseases, or vestibular or visual disturbances. Each subject was informed of the aims and protocols of this experiment and submitted informed consent before participation. The study protocol was approved by the Institutional Review Board of Peking Union Medical College Hospital, Beijing, China.
Equipment
A force plate (1600 Hz, AMTI, Watertown, MA, USA) was utilized to measure the vertical ground reaction force (vGRF). The force plate and surrounding floor had similar surface properties to avoid any potential imbalances. A 3D motion capture system (200 Hz, Vicon, Oxford, UK) was utilized to obtain kinematic data. Reflective surface marker sets were tightly attached to the corresponding bony landmarks. Eight cameras (CMOS, Vicon, Oxford, UK) containing sensors recorded the entire simulated parachute jump in a half-squat posture.
Two commercially available ankle braces and CPAB were used in this experiment: an elastic ankle brace (AQ5261EA, Tokyo, Japan) and a semi-rigid ankle brace (LP787, Seattle, WA, USA). The elastic ankle brace body was composed of an ultra-thin material, the inner shell of which was constructed of high-elastic anti-skid mesh fabric, and the outer shell was constructed of high-elastic shock-absorbing foam. Two straps crossing from the planta in a figure-eight pattern were pressurized and fixed at the lateral and medial malleoli to strengthen ankle joint stability (Fig. 1a). The semi-rigid ankle brace contained a U-shaped semi-rigid metal spring functioning as a “hoop” at the lateral and medial malleoli (Fig. 1b).
Fabrication of CPAB
The CPAB body was composed of a sparsely porous honeycomb-like material called pique fabric (Uniform Hse, Hongkong, China). An elastic metal spring strip (tensile strength: 800 MPa; torsional strength: 28 times / 360°; elongation: 3%) was constructed on the inner side and adapted to the anatomical outline of the medial malleolus. Also, a special aluminum strip (tensile strength: ≥370 MPa; yield strength: ≥215 MPa; elongation: ≥12%) was constructed on the outer side and adapted to the anatomical outline of the lateral malleolus. The aluminum strip was coated with cotton foam (100% ethylene-vinyl acetate copolymer) to reduce friction between the ankle joint and CPAB. The strips were the most novel part of the CPAB design, and their length was 15 cm, their width was 1.2 cm, and their thickness was 0.2cm. Another special part was the heel pad, which was made of the auxetic material with a negative Poisson ratio (polymer porous polytetrafluoroethylene). When the feet and ankle are impacted by huge ground reaction forces (GRF), the heel pad can resist excessive deformation and increase comfort. Moreover, impact load conduction is increased and instantaneous impact force is weakened. CPAB is pressurized by an adhesive band at the top and two crossed bands at the back of the foot, which users can utilize adjust the tightness and thereby strengthen the ankle joint stability (Fig. 1c-e).
Procedure
Before jumping, each subject jogged for 5 min at a comfortable speed as a warm-up, then performed the HSPL. Upon hearing the order to jump, the subject jumped forward and flexed their lower limbs with their knees, ankles, and forefeet hugging each other and with the plantar parallel to the ground. This was called “three huggings and one parallel” in the teaching material of the China Airborne School. Then they landed on the force plate until their trunk stopped moving and resumed a neutral stance [2]. Subjects were evaluated under four different ankle brace conditions (no brace, elastic ankle brace, semi-rigid ankle brace, and CPAB) and instructed to start and terminate the drop landing movement in a standing position, to jump off and touch down with both feet, to lean forward with the body while jumping, and finally to stop the fall smoothly in a half-squat position. Each subject performed this maneuver from three different heights (low: 40 cm, medium: 80 cm, and high: 120 cm), undergoing five trials under each condition. The experimental condition order was random to prevent any order effects. Any fatigue effects were mitigated by resting for at least a 60 s interval between landings under each condition.
Each subject landed on the force plate, which collected GRF signals. A 3D motion capture system was utilized to measure the 3D position of reflective markers in a global reference frame. Reflective markers were utilized to determine the positions of the bony landmarks as virtual dots. All bony landmarks were defined as a visual 3D model and analyzed with the Vicon Nexus 2.6 software (C-Motion Inc., Germantown, MD, USA), which was utilized to compute 3D kinematic variables, and the AnyBody model was built to conduct reverse dynamics analysis. All subjects were briefly asked the same questions after participation, including questions regarding ease of use, quality, comfort, stability, hindrance, and satisfaction. The multiple 5-point Likert scale was evaluated by the subjects with 5 being the best and 1 being the worst.
Data Collection and Statistical Analysis
(1) The kinetic parameters: GRF data were measured in the dominant foot. All vGRF values were normalized to body weight (BW) and the time to peak vGRF (T-PvGRF) started from initial contact with the force plate. The reverse dynamic variables included the maximal plantarflexion moment (MPM) and maximal eversion moment (MEM). (2) The kinematics parameters: angular displacement of maximal dorsiflexion (MDAD), the angular displacement of maximal inversion (MIAD), the angular velocity of maximal dorsiflexion (MDAV), and angular velocity of maximal inversion (MIAV), were calculated with the software (3). The energy parameters: the work and maximum power. From the mechanics perspective, the work refers to the amount of joint power conducted in a certain period time [10], and the calculation formula is as follows:

P represents joint power, and t1 and t2 represent the start and end time points. The work done by the ankle joint from initial contact with the force plate to buffering completion is negative work, namely energy absorption. Joint power = angular velocity × joint moment, which can be obtained by Anybody reverse dynamics analysis. In this study, dorsiflexion and inversion were stipulated to be positive, while plantarflexion and eversion were negative.
Data are representative of these experiments and are shown as the means ± standard deviation (SD). Two treatment groups were compared via the t-test of Students. Multiple group comparisons were performed via a two-way analysis of variance with Tukey’s post hoc test. Statistical analysis was conducted using GraphPad Prism 7.0 software, and statistical significance was declared as P < 0.05.