Effects of compression legwear on knee biomechanics and inter-joint coordination during depth jumps: a randomized cross-over study

Abstract

This study aimed to investigate the impact of different types of compression legwear on knee biomechanics, and lower limb inter-joint coordination during depth jumps. Twelve healthy male participants were randomly assigned to wear compression shorts (CS), compression tights (CT), and sports shorts (control: CC). The participants perform depth jumps from a platform of 40 cm in height, thus landing on their dominant side (DS) and non-dominant side (NS). The lower limbs kinematics and kinetics are simultaneously recorded by using a motion capture system and force plates. During initial contact phase, the CT group shows larger peak knee flexion angles (P_I = 0.033) and sagittal knee range of motion (ROM) (P_III = 0.010) but smaller horizontal ROM of the knee (P_III = 0.016) than the CC group. During stabilization phase, the CS group exhibits significantly higher peak knee flexion (P_II = 0.025) and sagittal knee ROM (P_II = 0.001). While the peak vertical ground reaction force (vGRF) is not obtained by the type of legwear (P > 0.05), significant phase- and direction-specific differences emerge in the knee moments, with the CS group showing larger extension moments in landings on the DS (P_I < 0.001, P_II = 0.005) but smaller abduction moments than the CC/CT in landings on the NS (P_II = 0.004). An inter-joint coordination analysis indicates that the CT group has the highest knee-ankle (P < 0.001) and knee-hip coordination (P < 0.001) during landing on the NS, whereas the CC group shows excellent knee-hip coordination on the DS (P < 0.001). Compression legwear exerts pressure on the lower limbs during various phases of jumping, thereby altering the biomechanical characteristics of the knee joint and the synergistic mechanisms of the lower limbs. CT improve knee-ankle coordination and mediolateral stability, while CS enhance sagittal-plane control during bilateral landings.

Introduction

Compression legwear is an important garment in modern sports and widely worn during athletic and leisure training due to its ability to enhance sports performance and reduce the risks of injury^1,2. This garment enhances the circulation of blood, accelerates post-exercise recovery, and significantly contributes to sports efficiency^3,4. Recent advancements in sports biomechanics have shifted from the focus on general functionality to refining the biomechanical mechanisms. For example, studies on vertical jumps show that compression legwear restricts lateral muscle vibrations⁵, thus directing more energy toward the generation of vertical forces thereby improving the jump height. This redirection of energy potentially enhances jump efficiency by reducing muscle oscillation losses. This effect is attributed to the pressure distribution of the legwear, which enhances lower limb muscle stability and joint coordination, thus leading to more precise neuromuscular control during movement⁶. Additionally, compression-induced improvements in proprioception help athletes to maintain better posture and movement trajectories, thus further optimizing performance⁷.

Among the many dynamic movements, jumping and landing are particularly demanding on the lower limbs, with the knee joint being the most vulnerable to injury due to the high impact forces experienced during take-off and landing⁸. Leabeater et al.⁹ found that compression legwear helps to control the lower limb muscles, thus promoting smoother knee joint movements, and preventing excessive knee extension or internal/external rotation, thereby mitigating knee injury risks. Furthermore, compression legwear enhances the overall stability of the body, which allows athletes to maintain better control during explosive movements such as jumps¹⁰. Zamporri and Aguinaldo¹¹ and Donath and Faude¹² suggested that compression legwear improves the dynamic stability of the knee joint, while dissipating the forces of the landing impact, and reducing strain on the soft tissues. Proper pressure from compression legwear can enhance proprioception¹³. However, excessive pressure can impair performance by restricting muscle elasticity and the ROM^14,15. Leabeater et al.⁹ reported that there is no statistically significant differences on jumping performance with the use of compression legwear, this inconsistency suggests that the effectiveness of compression legwear depends on optimizing its design parameters. Thus, effective pressure distribution is crucial for maximizing benefits without preventing athletic performance.

Beyond stability and injury prevention, compression legwear may also influence the fundamental neuromuscular control mechanisms that underlie athletic performance, particularly inter-segment coordination. The mastery of complex motor skills requires effective inter-segment coordination, which is defined as the synchronized movement of one joint relative to another^16,17. The lower limb coordination directly influences jumping performance¹⁸. These coordination patterns can be described and quantified by using a continuous relative phase (CRP) analysis^19,20. A CRP analysis is a practical approach for evaluating coordination by analyzing the spatial and temporal relationships between joints, thus offering valuable insights into movement efficiency²¹.Therefore, investigating hip-knee-ankle coordination under different compression conditions could help elucidate how compression legwear enhances biomechanical efficiency.

Despite the widespread use of compression legwear for performance enhancement and injury prevention, the aforementioned mixed biomechanical outcomes might be attributed to methodological differences, including different lengths and types of compression legwear used. Critically, few studies have systematically examined the effect of compression legwear on lower limb coordination during dynamic tasks such as jumping. Moreover, to the best of the knowledge of the authors, no previous study has investigated its effect on lower limb coordination during jumping. To address the limited understanding in this area, this study has conducted a comparative biomechanical analysis of the lower limb kinematics, kinetics, and joint coordination during depth jumps with different types of compression legwear (i.e., legwear that is ankle-length or above-knee length cut). By examining how different types of compression legwear affect movement mechanics, this research provides evidence-based recommendations for the practical use of compression legwear in sports and leisure training, which ultimately contributes to improved performance and injury prevention.

Method

Participants

The sample size for the experiment was calculated by using G*Power 3.1 (Power = 80% ; α = 0.05), and the minimum required number of participants is 10²². Therefore, 12 active male students were recruited from the university of the second author who regularly play basketball or badminton. The basic profile (mean (SD)) of the participants is as follows: age (25.0 ± 3.3 years old); height (172.0 ± 5.8 cm); body weight (67.6 ± 8.1 kg); and years of training (6.4 ± 5.0 years). All of the participants completed a questionnaire survey before the experiment, with inclusion criteria of no major lower limb injuries in the past 6 months, good physical condition, and athletic ability²³. In this study, the dominant leg was determined by kicking a ball²⁴. This study also received human subject ethics approval from the Institutional Review Board of the University Ethics Committee of the university of the second author. All of the participants signed informed consent forms prior to participating in the experiment.

Legwear

Two types of compression legwear of an international brand (2XU, Australia) are selected for this study. The first type is compression shorts (CS), which covers the area above the knee or legwear with an above-knee length cut; the other type is compression tights (CT), which cover the joints both above and below the knees with an ankle-length cut (see Fig. 1). The front panel of both CS and CT is made from 72% nylon and 28% polyurethane while the back panel is made from 65% nylon and 35% polyurethane. The control condition (CC) is based on the use of commercially available every day shorts, which are mainly made from polyester with low elasticity. According to industrial-provided compression ratings, CC induces 0–5 mmHg, while CS and CT induce 20–30 mmHg and 23–25 mmHg, respectively of pressure. The targeted muscle groups that are covered by CS and CC are the gluteus maximus, hamstring and vasti muscles, while the targeted muscle groups covered by CT are the gluteus maximus, hamstring, vasti, gastrocnemius, and soleus muscles. To ensure that the induced compression was consistent across all of the subjects, the prescription process for the legwear included individual measurements and fitting to assure the level of compression experienced by each subject.

Fig. 1.

Three different types of legwear.

Experiment protocol

This study uses a repeated measures design, with each participant wearing the three different types of legwear (i.e., CS, CT, and CC) randomly and performing multiple depth jumps. The participants were given detailed instructions on the experimental procedures and practiced the required movements. Before the data were formally collected, all of the participants took part in warm-up exercises under the guidance of a fitness coach. They completed three warm-up exercises, and performed two sets of each exercise. After the warm-up, they rested for 5–10 min. The testing sequence for the NS and DS, and legwear conditions was randomized by using a computerized randomization generator²⁵. To eliminate any potential impact from footwear, all of the participants were required to complete the depth jumps barefoot. A total of 30 reflective markers were attached to the lower limbs according to the Vicon Plug-in Gait model with additional tracking points (Fig. 2a)⁸. Marker placement followed standardized biomechanical protocols while accounting for individual anatomical variations.

Fig. 2.

Experimental data collection.

The participants completed 5 repetitions on each leg with a 30-second interval in between each repetition²⁶. A 15-minute interval was also provided to reduce fatigue between the different legwear conditions²⁷. During the depth jumps, the participants were instructed to place their hands on the iliac crest²⁸, and visually focus on a target on the wall in front of them, step forward with either NS or DS (rather than jumping down), landing on both feet. After landing, they were required to rebound jumps as fast as they could, then landed on the force plates and be stable for at least 3 s (Fig. 2b). A certified fitness coach monitored the quality of the movement, trials were considered unsuccessful if the participant landed out of force plates, or failed to remain stationary for the required 3 s, or removed his hands from his iliac crest during depth jump.

To minimize interference from the experimental equipment, a customized wooden box [60 cm (length) × 50 cm (width) × 40 cm (height)] was used as a jumping platform. This platform served as the take-off point for the jumps, which standardized the jump height²⁴. A nine-camera motion capture system (VICON, Nexus 2.0 Inc., UK, 100 Hz ) was used to collect the kinematic data^29,30. Two force plates that are 60 cm × 40 cm in dimensions (AMTI, USA, 1000 Hz) were used to collect the kinetic variables. The Vicon and AMTI systems were synchronized by using a digital signal converter. Static calibration trials were first acquired with the participants standing barefoot on two separate force plates, and maintaining a standard foot position.

Phase deviation

The kinematic phase definitions were determined by using the vGRF thresholds in Visual3D software (Dolomite Enterprises, LLC, USA). The touchdown moment (T1) was identified as the first frame where the vGRF exceeded 10 N following landing, while the take-off moment (T2) was marked when the vGRF fell below 10 N prior to take-off³¹. As shown in Fig. 3, the depth jump cycle comprises three distinct phases: (1) T1–T2 (initial contact phase) which encompasses the entire ground contact period from initial touchdown through to the landing or amortization phase then to propulsion at take-off; (2) T2–T3 or the airborne phase which represents the period of flight between take-off and the subsequent touchdown (T3); and (3) T3–T4 (stabilization phase), which represents the period from the second landing and cushioning to complete body stabilization (T4: with vertical ground reaction force variation on the single side within 5%).

Fig. 3.

Corresponding time points of event deviation.

Biomechanical indicators

Visual3D software was used to establish a lower limb skeletal model to calculate the kinematic and kinetic variable. All data undergoes filtering using a 4th-order Butterworth low-pass filter. The cutoff frequency for kinematic data is 15 Hz, while the cutoff frequency for GRF data is 55 Hz. The former includes jump height, peak knee flexion angle, and the ROM of the knee joint from three planes of motion (i.e., coronal, sagittal, and horizontal) were collected and analyzed following previous study⁸. Jump height was calculated based on the subject’s flight time. The latter include the vGRF, and peak knee joint moment. Both the peak vGRF and peak knee joint moment are standardized by body weight (BW). Joint coordination was calculated by using a CRP analysis, which was used to analyze the knee-ankle, and knee-hip joint coordination during the depth jump, the technical details are provided in previous publication³². A CRP value close to the zero-value line indicates higher coordination³².

Statistics

All of the experimentally measured data are presented as mean ± standard deviation (SD), and a statistical analysis was performed by using SPSS software (SPSS Statistics IBM, Version 26.0, USA). A one-way repeated measures analysis of variance (ANOVA) was used for the statistical analysis³³, and Mauchly’s test for sphericity was employed to evaluate whether the data met the sphericity assumption³⁴. If the assumption was not met, the Greenhouse-Geisser correction was used to adjust the F-value and degrees of freedom³⁴. Pairwise comparisons were conducted by using the least significant difference (LSD) method³⁴ to observe the effects of the different compression legwear on the biomechanical variables. The significance level is defined as α = 0.05³⁴.

Results

Kinematic outcomes

Table 1 presents the kinematic outcomes, including jump heights and knee joint variables, and the three types of legwear when landing on the NS/DS are compared across two different phases (T1–T2 and T3 – T4).

Table 1.

Knee joint kinematic variables across T1–T2 and T3–T4 (mean (SD)).

Item		CS (compression shorts)	CT (compression tights)	CC (control condition)	F value	P value	η 2	LSD results
								P I (CS vs. CT)	P II (CS vs. CC)	P III (CT vs. CC)
Jump height (cm)	NS	31.09 ± 6.72	32.23 ± 6.73	29.59 ± 6.26	1.228	0.273	0.058	0.277	0.392	0.175
	DS	31.25 ± 6.08	31.48 ± 6.03	28.94 ± 5.20	1.674	0.200	0.075	0.879	0.073	0.084
T1–T2	NS
	Peak knee flexion angle	101.76 ± 11.43	105.06 ± 12.41	99.07 ± 8.23	2.970	0.062	0.127	0.033*	0.421	0.054
	Sagittal ROM	100.60 ± 11.81	103.14 ± 11.54	99.95 ± 10.61	2.118	0.101	0.108	0.126	0.711	0.044*
	Coronal ROM	8.99 ± 2.47	8.88 ± 3.00	9.26 ± 3.33	0.273	0.762	0.007	0.826	0.615	0.498
	Horizontal ROM	8.04 ± 2.32	7.25 ± 2.34	7.59 ± 2.37	2.083	0.131	0.048	0.031*	0.247	0.430
	DS
	Peak knee flexion angle	100.30 ± 10.26	103.21 ± 12.45	100.20 ± 10.66	0.456	0.637	0.021	0.421	0.615	0.340
	Sagittal ROM	99.17 ± 10.90	102.21 ± 10.85	101.06 ± 9.35	2.214	0.115	0.049	0.058	0.184	0.426
	Coronal ROM	10.33 ± 3.87	11.02 ± 2.75	10.33 ± 2.75	3.513	0.039*	0.143	0.114	0.987	0.010*
	Horizontal ROM	7.81 ± 1.94	7.86 ± 1.59	8.56 ± 2.24	3.079	0.057	0.128	0.899	0.094	0.016*
T3–T4	NS
	Peak knee flexion angle	76.13 ± 15.60	74.34 ± 17.26	68.36 ± 14.70	2.948	0.063	0.123	0.337	0.025*	0.115
	Sagittal ROM	30.74 ± 8.61	32.92 ± 9.26	31.34 ± 8.87	1.380	0.258	0.310	0.163	0.587	0.274
	Coronal ROM	5.61 ± 4.24	4.74 ± 2.10	5.52 ± 1.65	6.340	0.003*	0.129	0.656	0.001*	0.012*
	Horizontal ROM	4.69 ± 2.24	3.90 ± 1.16	5.36 ± 1.83	9.563	0.000*	0.182	0.032*	0.048*	0.000*
	DS
	Peak knee flexion angle	77.14 ± 14.29	75.28 ± 17.74	69.41 ± 13.48	3.827	0.030*	0.154	0.338	0.015*	0.124
	Sagittal ROM	32.07 ± 7.41	33.40 ± 9.12	31.94 ± 8.57	0.475	0.625	0.022	0.399	0.909	0.341
	Coronal ROM	6.49 ± 2.34	6.57 ± 2.89	6.13 ± 1.87	0.706	0.496	0.016	0.848	0.304	0.299
	Horizontal ROM	5.33 ± 1.67	4.32 ± 1.54	5.60 ± 1.48	9.515	0.000*	0.181	0.003*	0.362	0.000*

Jump height (cm), Peak knee flexion angle (degree), Sagittal ROM (flexion/extension) (degree), Coronal ROM (abduction/adduction) (degree), Horizontal ROM (internal/external) (degree). *Indicates a significant difference.

No significant main effect of the three types of legwear can be observed for the jump height when landing on both the NS (P > 0.05, η² = 0.058) and DS (P > 0.05, η² = 0.075) with post-hoc LSD tests confirming that there is no pairwise significance.

In the T1–T2 stage, the CT group exhibits a significantly larger peak knee flexion angle than the CS group (P_I = 0.033, η² = 0.127, 95% CI ( − 6.42, − 0.28)) for the first landing on the NS. There are significant differences in the coronal ROM between the CT and CC (P_III = 0.044, η² = 0.108, 95% CI (0.094, 6.28)) groups, and the horizontal ROM between the CS and CT (P_I = 0.031, η² = 0.048, 95% CI (0.075, 1.512)) groups, while no significance is found in the sagittal ROM (P > 0.05). After the first landing on the DS, the legwear has a significant main effect for the sagittal ROM (P = 0.039, η² = 0.143,), with the CT group showing a significantly larger ROM than the CC group (P_III = 0.010, η² = 0.143, 95% CI (0.17, 1.21)). On the other hand, the CT group shows a significantly smaller horizontal ROM than the CC group (P_III = 0.016, η² = 0.128, 95% CI ( − 1.26, − 0.14)).

During T3–T4, the CS group shows a significantly larger peak knee flexion angle than the CC group (P_II = 0.025), η² = 0.123, 95% CI (1.00, 14.54 )) for the landing on the NS. For the sagittal ROM, significant group differences can be observed (P = 0.003, η² = 0.129), with the CS group exhibiting a significantly larger ROM than the CC group (P_II = 0.001, η² = 0.129, 95% CI (− 1.49, − 0.43 )), while the CT group exhibits a significantly smaller ROM than the CC group (P_III = 0.012, η² = 0.129, 95% CI (− 1.46, − 0.19)). The horizontal ROM is highly significant across all groups (P < 0.001, η² = 0.182). The CC group exhibits a significantly larger horizontal ROM than both the CS (P_II = 0.048, η² = 0.182, 95% CI (− 1.33, − 0.01)) and CT (P_III < 0.001, η² = 0.182, 95% CI (− 2.10, − 0.82 )) groups, while the CS group exhibits a significantly larger horizontal ROM than the CT (P_I = 0.032, η² = 0.182, 95% CI (0.07, 1.51 )) group. Similarly, the CC and CS groups both exhibit a larger horizontal ROM on the DS than the CT group (P_III < 0.001, η² = 0.181, 95% CI (0.37, 1.63 )and P_I = 0.003, η² = 0.181, 95% CI (− 1.90, − 0.63 )).

Kinetic outcomes

Table 2 presents the kinetic outcomes, including the peak vGRF values, and peak knee joint moments in the coronal, sagittal, and vertical axes, and compares the three types of legwear when landing on the NS/DS across T1–T2 and T3 – T4.

Table 2.

Knee joint kinetic variables across T1–T2 and T3–T4 (mean (SD)).

Item			CS (compression shorts)	CT (compression tights)	CC (control condition)	F value	P value	η 2	LSD results
									P I (CS vs. CT)	P II (CS vs. CC)	P III (CT vs. CC)
T1–T2	NS
	Peak vGRF		1.79 ± 0.42	1.73 ± 0.46	1.93 ± 0.50	1.133	0.332	0.054	0.826	0.201	0.118
	Peak knee joint moment	Flexion	0.67 ± 0.52	0.75 ± 0.68	0.60 ± 0.42	0.962	0.391	0.046	0.280	0.344	0.168
		Extension	2.20 ± 0.50	2.11 ± 0.61	2.07 ± 0.53	0.672	0.514	0.016	0.297	0.278	0.864
		Adduction	0.50 ± 0.87	0.82 ± 1.23	0.57 ± 1.17	2.464	0.098	0.110	0.030*	0.580	0.143
		Abduction	0.49 ± 0.17	0.43 ± 0.23	0.55 ± 0.21	3.928	0.024*	0.098	0.174	0.125	0.014*
		Internal	0.36 ± 0.11	0.32 ± 0.13	0.37 ± 0.13	1.997	0.143	0.051	0.126	0.671	0.089
		External	0.44 ± 0.77	0.61 ± 0.80	0.50 ± 1.00	1.563	0.222	0.720	0.086	0.557	0.388
	DS
	Peak vGRF		1.94 ± 0.47	1.82 ± 0.49	2.04 ± 0.47	1.291	0.286	0.058	0.430	0.183	0.150
	Peak knee joint moment	Flexion	0.82 ± 0.71	0.78 ± 0.54	0.67 ± 0.32	1.185	0.169	0.041	0.632	0.096	0.122
		Extension	2.25 ± 0.48	1.99 ± 0.47	2.06 ± 0.41	8.747	0.001*	0.176	0.000*	0.005*	0.321
		Adduction	0.61 ± 0.81	0.66 ± 0.80	0.66 ± 1.08	0.355	0.703	0.017	0.638	0.461	0.973
		Abduction	0.49 ± 0.21	0.50 ± 0.24	0.48 ± 0.14	0.255	0.776	0.007	0.797	0.600	0.510
		Internal	0.30 ± 0.18	0.32 ± 0.20	0.32 ± 0.18	0.279	0.758	0.013	0.631	0.494	0.967
		External	0.48 ± 0.63	0.54 ± 0.60	0.70 ± 1.07	4.614	0.015*	0.180	0.483	0.007*	0.218
T3–T4	NS
	Peak vGRF		2.02 ± 0.40	2.10 ± 0.44	2.16 ± 0.44	3.024	0.059	0.126	0.234	0.025*	0.258
	Peak knee joint moment	Flexion	0.48 ± 0.36	0.46 ± 0.45	0.57 ± 0.26	0.691	0.507	0.037	0.812	0.382	0.348
		Extension	2.28 ± 0.67	2.27 ± 1.27	2.44 ± 0.96	0.308	0.738	0.015	0.896	0.446	0.537
		Adduction	0.78 ± 1.49	1.17 ± 1.65	0.82 ± 1.60	0.812	0.455	0.061	0.258	0.790	0.207
		Abduction	0.42 ± 0.15	0.46 ± 0.20	0.53 ± 0.14	4.143	0.020*	0.118	0.029*	0.004*	0.941
		Internal	0.33 ± 0.11	0.32 ± 0.14	0.37 ± 0.11	1.339	0.269	0.039	0.972	0.120	0.186
		External	0.57 ± 0.88	0.79 ± 0.87	0.66 ± 1.03	0.677	0.514	0.031	0.264	0.505	0.483
	DS
	Peak vGRF		2.21 ± 0.54	2.19 ± 0.58	2.17 ± 0.49	0.022	0.790	0.001	0.817	0.875	0.105
	Peak knee joint moment	Flexion	0.59 ± 0.35	0.52 ± 0.44	0.73 ± 0.27	2.066	0.140	0.094	0.463	0.180	0.099
		Extension	2.42 ± 0.69	2.37 ± 1.03	2.24 ± 0.41	2.193	0.125	0.099	0.610	0.052	0.355
		Aadduction	0.44 ± 0.67	0.95 ± 1.30	0.41 ± 0.75	2.550	0.095	0.145	0.029*	0.747	0.040*
		Abduction	0.53 ± 0.29	0.55 ± 0.22	0.46 ± 0.15	3.689	0.037*	0.197	0.772	0.149	0.035*
		Internal	0.42 ± 0.16	0.43 ± 0.14	0.41 ± 0.16	0.030	0.971	0.001	0.839	0.968	0.830
		External	0.60 ± 0.07	0.80 ± 0.13	0.56 ± 0.10	1.120	0.336	0.053	0.212	0.484	0.143

Peak vGRF (BW), Peak knee joint moment (N m / kg), *Indicates a significant difference.

During T1–T2, the legwear has no significant main effect on the peak vGRF value (P = 0.332, η² = 0.054), with no significance across all groups (P > 0.05). For the first landing pattern on the NS, the legwear only has a significant main effect on the peak knee abduction moment (P = 0.024, η² = 0.098). Post-hoc LSD tests reveal that the CC group shows a significantly larger peak knee abduction moment than the CT group (P_III = 0.014, η² = 0.098, 95% CI (0.03, 0.21)). Although the legwear has no significant main effect on the peak knee adduction moment (P = 0.098, η² = 0.110), the LSD test shows that the CT group exhibits a significantly larger peak knee adduction moment than the CS group (P_I = 0.030, η² = 0.110, 95% CI (− 0.60, − 0.03 )). For the first landing pattern on the DS, the legwear has a significant main effect on the peak knee extension moment (P < 0.001, η² = 0.176), with the CS group showing a significantly larger peak knee extension moment than both the CT (P_I < 0.001, η² = 0.176, 95% CI ( 0.14, 0.41)) and CC (P_II = 0.005, η² = 0.176, 95% CI(0.07, 0.34 )) groups. Additionally, the legwear has a significant main effect on the peak knee external moment (P = 0.015, η² = 0.180), with the CC group showing a significantly larger peak knee external moment than the CS group (P_II = 0.007, η² = 0.180, 95% CI (0.06, 0.38)).

In the T3–T4 phase, the legwear has no significant main effects on the peak vGRF in the landing pattern on both the NS and DS (P > 0.05). However, the CC group has a significantly higher peak vGRF than the CS group (P_II =0.025 0.05, η² = 0.126, 95% CI (− 3.6, − 0.03 )) during landing on the NS. For the peak knee abduction moment, the legwear has significant main effects on the landing patterns of both the NS (P = 0.020, η² = 0.118) and DS (P = 0.037, η² = 0.197). During landing on the NS, the CS group shows a significantly smaller peak knee abduction moment than both the CT (P_I = 0.029, η² = 0.118, 95% CI (0.01, 0.20)) and CC (P_II = 0.004, η² = 0.118, 95% CI (0.04, 0.17)) groups. During landing on the DS, the CT group shows a significantly larger peak knee abduction moment than the CC group (P_III = 0.035, η² = 0.197, 95% CI (− 0.16, − 0.01)). Although the legwear has no significant main effect on the peak knee adduction moment (P = 0.095, η² = 0.145), the CT group shows a significantly larger peak knee adduction moment than the CS (P_II = 0.029, η² = 0.145, 95% CI (− 0.96, − 0.06)) and CC (P_III = 0.040, η² = 0.145, 95% CI (0.03, 1.05)) groups.

Inter-joint coordination outcomes

Figure 4 depicts the inter-joint coordination during T1–T2. The legwear has no significant main effect on the knee-ankle coordination of both the NS (P = 0.070, η² = 0.053) and DS (P = 0.121, η² = 0.042). However, significance is observed between the CS and CT groups (P_I = 0.040, η² = 0.053, 95% CI (0.03, 1.39)) for landing on the NS, with the CRP curve of the CT group closer to the zero value line. There is a significant difference between the CS and CC (P_II = 0.040, η² = 0.042, 95% CI (0.05,1.89)) groups for landing on the DS, with the CRP curve of the CS closer to the zero value line. For knee-hip coordination, the legwear only has a significant main effect on landing on the NS (P = 0.002, η² = 0.122), with the CRP curve of the CC group closer to the zero value line, which shows a significantly higher knee-hip joint coordination than the CS (P_II = 0.042, η² = 0.122, 95% CI (− 1.14, − 0.02)) and CT (P_III < 0.001, η² = 0.122, 95% CI (− 1.19, − 0.35)) groups.

Fig. 4.

Plotted lower limb joint coordination during T1–T2.

Figure 5 depicts the inter-joint coordination during T3–T4. For the knee-ankle coordination, the legwear has a significant main effect for landing on both the NS (P < 0.001, η² = 0.476) and DS (P < 0.001, η² = 0.464). For landing on the NS, highly significant differences are observed across all of the groups (P_I < 0.000, η² = 0.476, 95% CI (− 7.45, − 4.58 ). P_II = 0.002, η² = 0.476, 95% CI (0.64, 2.91). P_III < 0.000, η² = 0.476, 95% CI (6.11, 9.46)). The CS group shows optimal knee-ankle joint coordination from touchdown to the maximum knee flexion moment with the CRP value closer to the zero value line, while the CT group shows optimal knee-ankle joint coordination from the maximum knee flexion to the standing moment. For landing on the DS, significances can be observed across all of the groups, with the CC group exhibiting optimal knee-ankle joint coordination during touchdown and the maximum knee flexion moment. After the maximum knee flexion moment, the CT group shows optimal knee-ankle joint coordination.

Fig. 5.

Plotted lower limb joint coordination during T3–T4.

For knee-hip joint coordination, the legwear has a significant main effect on both the NS (P < 0.001, η² = 0.215) and DS (P < 0.001, η² = 0.365). For landing on the NS, the CT group exhibits a significantly optimal knee-hip joint coordination (P_I < 0.001, η² = 0.215, 95% CI (1.73, 3.94). P_III = 0.012, η² = 0.215, 95% CI (− 3.40, − 0.42)). On the contrary, the CC group exhibits optimal knee-hip joint coordination (P_II < 0.001, η² = 0.365, 95% CI (− 3.27, − 1.50). P_III < 0.001, η² = 0.365, 95% CI (4.58, 8.20)) when landing on the DS.

Discussion

The key findings of this study are the three types of legwear significantly influence the knee kinematics and kinetics. During the T1–T2 phase, although the legwear types showed no significant effects on jump height or most kinematic and kinetic variables. Although the mean values suggested that the CT group had the highest jump height, followed by the CS group, with the CC group lowest, the main effect of compression type on jump height was not statistically significant. Its practical effect could not be established. In the T3–T4 phase, significant differences across legwear types emerged in joint angles, ROM, and coordination patterns, with the CS and CT groups exhibiting more favorable biomechanical profiles—particularly in joint coordination and reduced horizontal ROM. The effects of compression legwear on knee biomechanical variables appeared to differ between the DS and NS sides. However, because this study was not specifically designed to examine limb asymmetry, these findings should be interpreted with caution. One plausible explanation is inherent neuromuscular asymmetry in landing and take‑off strategies.

Kinematics

The knee joint flexion angle plays a critical role in movement stability and control during depth jumps. Our findings reveal distinct kinematic differences among the different types of legwear, particularly in terms of knee flexion and ROM. The CT group shows a significantly larger peak knee flexion angle on the NS during T1–T2, though the actual difference was small, and its practical significance warrants further consideration. Which may have been attributed to the extended pressure, with applied pressure from the thigh to ankle, thus promoting greater knee flexion to optimize load absorption upon landing³⁵. However, significant differences were observed across all three planes of motion, indicating that compression legwear influences multiplanar knee kinematics. Nevertheless, the differences in the sagittal plane were relatively small, with more pronounced changes occurring in the frontal and transverse planes. This aligns with the natural biomechanics of jumping, where the sagittal plane is the primary plane of force generation and knee motion is more stereotypical in flexion and extension³⁶. In the T3–T4 phase, the CS group shows the largest peak knee flexion angles for landings on both the DS and NS. This indicates that the localized compression on the thigh of the CS group may enhance proprioceptive feedback, future studies incorporating direct measures of proprioception or muscle activation are warranted to verify this potential mechanism³⁷. However, the CC group shows a larger horizontal ROM at the knee, which suggests compensatory reliance on the ligamentous and muscular tension for stability. While this may improve lateral mobility, it could also reduce sagittal-plane control, thus potentially increasing the risk of ACL injury during high-impact landings³⁸.

The CT group shows reduced coronal and sagittal ROM on the NS during the depth jumps compared to the CS group, which implies that ankle-length compression leggings may enhance mediolateral stability for the less stable limbs¹⁰. However, this stabilizing effect is not found on the NS during the depth jumps, where the CT offer a larger ROM than the CS, which means that ankle-length compression leggings may restrict dynamic adaptability in the dominant limb during multidirectional movements³⁹. In contrast, the CC group consistently shows higher sagittal mobility bilaterally, thus suggesting unrestricted movement that relies solely on the passive ligamentous and muscular tension to regulate rotational control, a mechanism potentially associated with accelerated joint wear⁴⁰.

Kinetics

Different types of compression legwear have different impacts on the knee joint moment. During the T1–T2 phase, the CC group exhibits a larger peak knee abduction moment when landing on the NS, thus indicating that without external support, the neuromuscular system compensates by increasing abductor muscle (such as the gluteus medius, gluteus minimus, and iliotibial band help prevent harmful knee valgus) activation to maintain lateral stability. In contrast, the CT group shows a reduced peak knee abduction moment, thus suggesting that circumferential compression provides effective mediolateral stabilization. While the CT effectively reduce unwanted knee abduction moments, this type of legging also increases the knee adduction moment on the NS which leads to a higher medial load on the knee joint, which in turn, is detrimental to the medial compartment¹⁰.

The rotational kinetics reveals important differences, with the CC group showing a larger peak knee joint moment in the external direction during landing on the DS. This compensatory mechanism, while maintaining rotational control, may increase ACL shear forces⁴¹. These findings are consistent with recent studies^42–45, which shows that compression legwear can improve joint stability by applying external pressure and reducing abnormal joint moments, but may simultaneously reduce neuromuscular adaptation⁴². Thus, an optimal compression design must balance joint protection with performance maintenance.

While no significant between-group differences are found with the peak vGRF, the CC group exhibits higher peak vGRF values during landings on both the NS and DS. This shows that uncompressed limbs may use a more stiffer landing strategy, this may increase the impact force between the knee joint cartilage⁴¹. The lack of significant differences supports the findings of Lee et al.^10,46, thus indicating that compression garments primarily influence the joint mechanics rather than magnitude of the overall impact.

Inter-joint coordination

The CRP analysis reveals distinct coordination patterns between the different types of compression leggings across all phases. In T1–T2, the CT group shows superior knee-ankle coordination compared to the CS group with landing on the NS, as the CRP curve is the closest to the zero line. This suggests that ankle-length compression leggings promote both ankle plantarflexion and knee flexion at the same time^47,48. In contrast, the CS only apply pressure to the thigh muscles, and results in reduced knee-ankle coordination upon landing on the NS, as the lack of compression on the calves may decouple the distal-proximal joint synchronization. Notably, the CS group shows significantly higher ankle-knee coordination compared to the CC group during landing on the DS, thus indicating that a specific amount of compression induced onto the thigh muscles may enhance coordination in the stronger/dominant limb. However, the CC group maintains the best hip-knee coordination during landing on both the NS and DS. This finding suggests that while compression legwear can enhance specific joint couplings, they also simultaneously interfere with the natural inter-segmental coordination strategies of the body, particularly between the more proximal joints. This phenomenon could reflect a trade-off between external support and inherent neuromuscular control mechanisms, where uncompressed movement preserves the most physiologically coordinated patterns for certain pairs of joints³⁹.

In the T3–T4 phase, the effects of the compression legwear on joint coordination are highly phase-dependent. During the initial flexion phase, the CS group exhibits superior knee-ankle joint coordination during landing on the NS, while the CC group performs best on the DS landing, again highlighting the limb-specific responses to external support. However, following maximum knee flexion, the CT group shows the best knee-ankle coordination bilaterally, thus suggesting ankle-length compression enhances energy dissipation during weight acceptance. Hip-knee coordination reveals further complexities, with the CT group showing optimal coordination during landing on the DS, whereas the CC group maintains a higher degree of coordination during landing on the NS. A phase-specific analysis reveals the most synchronized patterns during simultaneous joint flexion for the CS group (NS) and CT group (DS). During extension, coordination dominance shifts to favor the NS of the CT group and the DS of the CC group.

These findings show that the effects of compression legwear on inter-joint coordination are highly dependent on a number of factors including: (1) specific coupled joint actions, (2) the phase of movement, (3) limb dominance, and (4) the extent of coverage of the compression garment. The results suggest that ankle-length compression legwear may be the most beneficial for energy absorption during bilateral landings, while above-knee compression legwear may offer advantages for single-leg stabilization tasks. However, the consistent superiority of uncompressed conditions for certain coupled joint actions indicates that the industry should carefully balance the benefits of external support against the potential disruption of natural coordination patterns. Future development of compression legwear should consider these nuanced, task-specific effects to optimize both performance and joint protection.

Limitations

This study has certain limitations. First, the pressure values of the compression legwear are provided by the manufacturer instead of directly measuring from the wearer. The exerted pressure from compression legwear can be vary due to individual differences. These differences could affect the lower limbs’ biomechanical changes. Second, while this study examines the impact of a small peak knee flexion angle on the biomechanics of the knee joint in the lower limb, the hip and ankle joints have not been fully explored. Furthermore, differences in muscle activity exert a significant influence on biomechanical variables; however, this study did not analyze the effects of compression legwear on muscle activity. Lastly, although the LSD test is useful for exploratory pairwise comparisons, it carries a higher risk of false positives and should be interpreted with caution.

Conclusions

This study shows that the compression legwear design significantly influences knee Joint biomechanics during depth jumps. It has certain positive effects on depth jumps performance and injury prevention in jumping movements, with distinct effects that depend on the garment length and landing phase. The key findings reveal that (1) CT improve knee-ankle coordination and mediolateral stability but increase medial knee loading; (2) CS enhance sagittal-plane control during bilateral landings; and (3) CC preserve natural hip-knee coordination while requiring greater neuromuscular compensation. The trade-offs among joint protection, coordination enhancement, and movement adaptability highlight the need for task-specific compression designs. Future research should standardize pressure measurements and expand investigations to interactions among multiple joints to optimize sport-specific applications. These findings suggest that selecting appropriate compression garments may improve movement efficiency and reduce injury risks associated with vertical jump landings, especially in sports training sport-specific applications and inform targeted prevention strategies.

Author contributions

Conceptualization, Q.Q.S., K.L.Y. and L.L.; Data curation, L.L., F.R.H., H.C.L.; Funding acquisition, Q.Q.S.; Investigation, Q.Q.S. and K.L.Y.; Methodology, Q.Q.S., K.L.Y., L.L., H.C.L., and W.S.M.; Supervision, Q.Q.S.; Writing—original draft, Q.Q.S., L.L., and F.R.H.; and Writing—reviewing & editing, QQS and KLY. All authors have read and agreed to the published version of the manuscript.

Funding

We acknowledge financial support from the Research Institute for Sports Science and Technology, The Hong Kong Polytechnic University (project code: 1-CD9Z) for this research project.

Data availability

The authors declare that the main data and materials that support the findings and conclusions of this study are available within the article.

Declarations

Competing interests

The authors declare no competing interests.

Ethical approval and consent to participate

This study received human subject ethics approval from the Institutional Review Board of The Hong Kong Polytechnic University Ethics Committee (HSEARS20240520004).

Informed consent

Statement: All of the study participants signed informed consent forms to participate in this study. Clinical trial number: not applicable.

Institutional review board statement

This study was approved by Hong Kong Polytechnic University, this research is carried out in accordance with relevant guidelines and regulations.