Effects of Unstable Exercise Using the Inertial Load of Water on Lower Extremity Kinematics and Center of Pressure During Stair Ambulation in Middle-Aged Women with Degenerative Knee Arthritis

Abstract

Stair ambulation requires precise lower extremity control and postural stability. Middle-aged women with degenerative knee arthritis (DKA) are at an increased risk of falls, yet the effects of unstable load training on their postural stability remain underexplored. This study investigated the effects of a 10-week Aqua Vest-based unstable load training program on postural stability and pain during stair ambulation in middle-aged women with DKA. Thirty participants were randomly assigned to an exercise group (EG) or a control group (CG), with 15 participants in each group. The EG completed a 10-week Aqua Vest training program, while the CG received no intervention. Pre- and post-assessments included spatiotemporal parameters, lower extremity kinematics, and center of pressure (CoP) trajectories during stair descent, as well as knee pain evaluated by a visual analog scale (VAS). After training, the EG demonstrated significant improvements in spatiotemporal and kinematic parameters, reduced ML displacement, lower VAS scores, as well as significant changes in AP CoP parameters. These findings suggest that Aqua Vest-based unstable load training may enhance ML postural stability and alleviate pain in DKA patients, potentially contributing to enhanced balance function and improved stair ambulation safety.

1. Introduction

Maintaining balance and preventing postural instability are critical concerns for middle-aged adults, as these factors are closely linked to an increased risk of falls in older populations [1,2]. Among middle-aged and older adults, degenerative knee arthritis (DKA)—the most common form of osteoarthritis (OA) affecting the knee joint—is a prevalent condition that leads to knee pain, altered gait mechanics, restricted joint motion, and increased joint loading, particularly during stair descent [3,4,5,6]. This progressive condition is characterized by cartilage degeneration, pain, and stiffness, all of which contribute to impaired mobility and gait dysfunction [7].

The prevalence of OA in middle-aged and older adults is influenced by various factors, including occupational activities, sports participation, musculoskeletal injuries, obesity, and gender [6,7,8]. Evidence suggests that older women are at a higher risk of falls and fall-related injuries than men, likely due to biomechanical constraints such as reduced sagittal plane moments, increased frontal plane hip joint moments, and diminished joint power during stair descent [7,8,9,10,11]. Center of pressure (CoP) movement in the anterior–posterior (AP) and medial–lateral (ML) directions is commonly used to assess balance and postural stability, with increased CoP displacement indicating compensatory control strategies [12,13]. Stair ambulation, a fundamental activity of daily living (ADL), imposes greater balance demands than level walking [14,15], requiring higher knee joint loading and an increased range of motion (ROM) in the sagittal plane during both ascent and descent [16,17]. Notably, stair descent presents even greater postural challenges, as it is associated with increased CoP sway, further elevating fall risk [18].

Several studies have established a strong link between DKA and fall risk, high-lighting the need for targeted interventions [8,19,20,21,22]. As DKA progresses, it leads to increasing mobility impairments, ultimately contributing to disability. Because knee OA is a progressive and irreversible condition, early-stage treatment typically focuses on conservative management strategies, including joint stress relief, rehabilitation, and physical therapy [5,6]. Various exercise-based interventions have been explored for DKA, including aerobic exercise, gait retraining, resistance training, and balance training. Zeng et al. [23] stated that low-intensity aerobic exercise is beneficial for severe knee OA patients, while high-intensity aerobic exercise is more suitable for mild knee OA patients with chronic conditions, promoting cartilage recovery, enhancing immunity, and alleviating pain. Khalaj et al. [24] emphasized the role of gait retraining in restoring normal gait mechanics and improving stability. Additionally, prior research has demonstrated that resistance training can partially restore functional deficits, balance impairments, and lower limb loading deficiencies in older women with knee OA [25,26]. Incorporating balance training has also been shown to enhance knee stability, reduce joint loading, alleviate pain, and improve overall functional performance [20,27].

Over the past decade, there has been growing interest in resistance training under unstable conditions, including the use of unstable surfaces or unstable load devices such as Swiss balls, BOSU balls, wobble boards, air cushions, slash pipes, and water- or sand-filled bags [28,29]. While research has primarily focused on unstable surface training, highlighting its benefits for muscle activation, postural stability, and proprioception [30,31], limited studies have explored the efficacy of unstable load training, particularly using water’s inertial properties. Studies suggest that unstable load training enhances stabilizer and antagonist muscle activation, improving upper body strength and joint stability more effectively than stable load training [32,33,34,35]. Water-filled resistance tools, such as training bags, have been reported to increase training intensity, elicit adaptive responses, and enhance performance [36]. Compared to traditional barbells, water-filled tubes induce greater core muscle activation and present a greater challenge to postural stability [1]. However, research on the application of unstable load-based functional training and rehabilitation remains scarce, particularly regarding its effects on postural stability in individuals with knee arthritis. Furthermore, most previous studies have focused on the short-term effects of various unstable devices, leaving a significant gap in understanding their long-term impacts. Therefore, evaluating the long-term effects of training methods that use water-based instability is crucial for advancing rehabilitation strategies.

To address this gap, the present study aims to evaluate the impact of a 10-week Aqua Vest-based unstable load training program on postural stability, lower extremity kinematics, and pain during stair ambulation in middle-aged women with DKA. We hypothesize that unstable exercise training with the Aqua Vest will alleviate knee pain while improving functional performance during stair negotiation. Findings from this study are expected to contribute to the optimization of fall prevention and rehabilitation programs for individuals with DKA.

2. Materials and Methods

2.1. Participants

Participants in this study were middle-aged women residing in Busan who voluntarily enrolled after responding to a recruitment announcement issued by Busan University of Foreign Studies’ Lifelong Education Center. The study was conducted from May 2023 to July 2023. To be eligible for inclusion, individuals needed to meet the following criteria: (1) be 45–60 years of age; (2) experience knee pain when completing activities of daily living (ADLs); (3) have knee pain for more than three months, with a visual analog scale (VAS) score of at least 30 mm; and (4) have a Knee Injury and Osteoarthritis Outcome Score (KOOS) below 80. All participants provided informed consent to participate in this study with the recommendation and approval of their physicians, and this study was performed in accordance with all ethical human research guidelines. Participants were excluded if they had any known neurological diseases or other conditions with the potential to impact typical locomotion. All participants signed the provided informed consent forms, which had received approval from the university ethics committee. The Institutional Review Board (IRB) also approved all study protocols.

This study enrolled a total of 30 participants (

Table 1. Participant characteristics (mean ± SD).

	EG (N = 15)	CG (N = 15)	p-Value
Age (yr)	55.73 ± 3.83	55.67 ± 3.16	0.959
Weight (kg)	58.87 ± 6.69	57.25 ± 5.69	0.483
Height (cm)	159.02 ± 4.12	156.75 ± 3.65	0.121
BMI (kg/m2)	0.23 ± 0.02	0.23 ± 0.02	0.993

) who were randomly assigned to the exercise group (EG) and the control group (CG) (N = 15/group). Individuals in the EG group completed a 10-week unstable exercise program using an Aqua Vest, whereas individuals in the CG group did not complete any specific exercise intervention.

2.2. Experimental Protocol

This study included three key stages, including (1) a pre-intervention test of stair ambulation, (2) a 10-week Aqua Vest training session, and (3) a post-intervention test of stair ambulation. For the pre- and post-intervention testing, all participants were asked to complete ascent at descent trials (N = 6 each) at their self-selected speed, including 3 trials each leading with the left leg and 3 leading with the right leg for both ascent and descent (N = 12 total trials). These stair ambulation tasks were completed with a customized four-step wooden staircase (height: 17 cm; depth: 28 cm) equipped with two embedded force plates (Figure 1). These AMTI-OR6 force platforms (AMTI, Watertown, MA, USA) were placed on the first and second steps and used to define initial contact and toe-off while also collecting vertical ground reaction force (VGRF) data for each foot (1000 Hz). Synchronized lower extremity kinematic data were collected with a seven-camera Vicon Nexus three-dimensional (3D) motion analysis system (Vicon Motion Systems Ltd., Oxford, UK) at a 100 Hz sampling frequency.

2.3. Aqua Vest Training Program

The 10-week intervention period was selected based on previous studies demonstrating that exercise programs lasting between 8 and 12 weeks effectively improve knee function and alleviate pain in individuals with knee osteoarthritis [37,38]. The study was conducted from May 2023 to July 2023. The 10-week Aqua Vest unstable exercise program was designed with the goal of enhancing postural control and stability of the lower extremities (see Figure 2). This program consisted of two weekly 5-minute sessions during which participants wore an Aqua Vest (Smartfits, Busan, Republic of Korea) containing water, providing an overall weight of 4–5 kg (see Figure 3). The selection of the Aqua Vest weight was informed by previous research, which recommended a consistent 5 kg load for water inertia-based training in middle-aged women, irrespective of exercise intensity [39]. In this study, the Aqua Vest weight was set between 4 and 5 kg, aiming to balance manageable weight with sufficient intensity for the exercise intervention. An adaptive approach was incorporated to accommodate individual comfort and prevent undue strain. For participants without shoulder discomfort, the weight was standardized at 5 kg, whereas for those experiencing shoulder discomfort, the load was reduced to ensure both comfort and effective participation.

These structures sessions consisted of a 5-minute warm-up period, followed by the following primary exercises (each lasting 5 min) targeting different facets of lower extremity balance and strength, including squats with lateral upper body swings (Session 1), forward lunges with arms raised (Session 2), lateral lunges with an Aqua Vest and a 10 kg barbell (Session 3), one leg stance and forward lean (Session 4), 3D lunges (Session 5), single-leg dynamic raises on a 30 cm plyo box (Session 6), wall touches with hip hinge using a 10 kg barbell (Session 7), and squat lunges with rotations on a 15 cm plyo box with an Aqua Vest (Session 8). After completing these exercises, each session concluded with 10 min of dynamic stretching using a foam roller. The Supplementary Materials provide additional details, including Table S1: Aqua Vest unstable exercise program; Video S1: Aqua Vest Training Program.

2.4. Data Analysis

Spatiotemporal, kinematic, and CoP (center of pressure) variables were assessed for each stair ascent and descent trial. Spatiotemporal parameters included stride time, step time, stride length, step length, and walking speed. Kinematic outcomes comprised hip, knee, and ankle joint angles in the frontal and sagittal planes. The gait cycle was defined as the period from the initial foot contact on the step with the embedded platform to the subsequent foot contact on the third step.

CoP measures were evaluated during the single-stance phase of the first step in stair descent. CoP trajectory was recorded using a force platform embedded in the second step, generating a time series of CoP positions in the AP and ML dimensions. CoP-related data were analyzed for the initial descent phase, from foot contact to toe-off on the force plate. Range and sway excursion of CoP displacement were calculated in both AP and ML directions during the single-stance phase. Mean CoP velocities in the AP and ML dimensions were determined using the first central difference method. CoP sway variability was assessed using root mean square (RMS) quantification in both dimensions.

Knee pain was assessed pre- and post-intervention using a VAS, represented by a 10 cm scale ranging from 0 to 100 mm. Participants were instructed to mark their perceived level of knee joint pain experienced during daily stair ambulation.

2.5. Statistical Analyses

All statistical analyses were performed using SPSS 26.0 (IBM, Armonk, NY, USA). Paired t-tests with Bonferroni correction were conducted to compare pre- and post-intervention spatiotemporal and kinematic variables. Independent samples t-tests were used to assess differences between the EG and CG groups, evaluating the effects of the 10-week Aqua Vest training program. Descriptive statistics were calculated for all experimental data. Normality of the data was assessed, and a significance threshold of p < 0.05 was applied.

3. Results

3.1. Knee Pain

A significant difference in VAS scores was observed between the EG and CG groups following the intervention (p < 0.001) (

Table 2. VAS comparisons within and between groups.

		EG (N = 15)	CG (N = 15)	t
VAS (cm)	Pre	4.33 ± 1.05	3.80 ± 1.01	−1.417
	Post	1.20 ± 1.03	4.27 ± 1.08	7.939 ***
	t	7.650 ***	−1.542

VAS, visual analog scale; *** p < 0.001. t: t-value from the independent t-test; t: t-value from the paired t-test.

). Specifically, the EG group exhibited a significant reduction in pain intensity after the 10-week exercise program (p < 0.001).

3.2. Spatiotemporal Parameters

Spatiotemporal gait parameters were measured before and after the exercise intervention for both groups (

Table 3. Spatiotemporal gait parameters while ascending and descending stairs (mean ± SD).

			Stair Ascent		Stair Descent
Variables		Training	EG	CG	EG	CG
Stride Time (s)	Non-dominant	pre	1.55 ± 0.24	1.57 ± 0.14	1.28 ± 0.25	1.28 ± 0.11
		post	1.35 ± 0.09 **, †	1.54 ± 0.14	1.17 ± 0.09 *, †	1.28 ± 0.12
	Dominant	pre	1.56 ± 0.24	1.57 ± 0.15	1.27 ± 0.24	1.28 ± 0.12
		post	1.36 ± 0.09 **, †	1.53 ± 0.14	1.16 ± 0.09 *, †	1.33 ± 0.19
Step Time (s)	Non-dominant	pre	0.79 ± 0.11	0.81 ± 0.07	0.63 ± 0.13	0.62 ± 0.06
		post	0.70 ± 0.04 **, †	0.78 ± 0.08	0.58 ± 0.04†	0.62 ± 0.06
	Dominant	pre	0.79 ± 0.13	0.79 ± 0.08	0.63 ± 0.12	0.62 ± 0.06
		post	0.69 ± 0.06 **, †	0.78 ± 0.08	0.58 ± 0.05 *, †	0.63 ± 0.05
Stride Length (m)	Non-dominant	pre	0.64 ± 0.02	0.64 ± 0.01	0.74 ± 0.04	0.74 ± 0.04
		post	0.64 ± 0.01	0.64 ± 0.01	0.77 ± 0.05 *	0.76 ± 0.07
	Dominant	pre	0.65 ± 0.01	0.64 ± 0.01	0.73 ± 0.03	0.74 ± 0.03
		post	0.64 ± 0.02	0.64 ± 0.01	0.77 ± 0.05 *, †	0.73 ± 0.04
Step Length (m)	Non-dominant	pre	0.32 ± 0.01	0.32 ± 0.01	0.43 ± 0.04	0.41 ± 0.04
		post	0.33 ± 0.01	0.33 ± 0.01	0.45 ± 0.05 *	0.43 ± 0.07
	Dominant	pre	0.33 ± 0.01	0.32 ± 0.01	0.41 ± 0.03	0.43 ± 0.04
		post	0.32 ± 0.01	0.33 ± 0.01	0.45 ± 0.05 *, †	0.41 ± 0.04
Walking Speed (m/s)	Non-dominant	pre	0.43 ± 0.07	0.41 ± 0.04	0.60 ± 0.11	0.58 ± 0.07
		post	0.48 ± 0.03 **, †	0.42 ± 0.04	0.66 ± 0.08 *, †	0.60 ± 0.08
	Dominant	pre	0.43 ± 0.06	0.41 ± 0.04	0.60 ± 0.11	0.59 ± 0.07
		post	0.47 ± 0.03 **, †	0.42 ± 0.03	0.67 ± 0.08 **, †	0.57 ± 0.10

* p < 0.05/** p < 0.01, pre- vs. post-training; † p < 0.05, EG vs. CG groups. All parameters differed significantly.

), revealing significant post-intervention differences between groups. Participants in the EG demonstrated a more efficient gait during stair ascent, characterized by reduced stride and step times compared to the CG (p < 0.05). Additionally, the EG exhibited significant increases in stride length (right: p < 0.05) and walking speed (left: p < 0.05, right: p < 0.05).

Within-group comparisons further indicated that EG participants experienced significant improvements in stair ascent performance, including reduced stride and step times, increased stride and step lengths, and enhanced walking speed (p < 0.01). However, no significant differences in stride or step length were observed within or between groups for stair descent.

Compared to the CG, individuals in the EG demonstrated significant improvements in stair descent parameters, including reduced stride time, step time, and increased walking speed (all p < 0.05). Within the EG, stride and step times were significantly reduced (p < 0.05), except for the left-side step time. Additionally, significant increases were observed in stride length, step length, and walking speed (p < 0.05), with the right leg showing more pronounced improvements in walking speed (p < 0.01). In contrast, the CG showed no significant changes in any gait parameters between pre- and post-intervention assessments.

3.3. CoP Measurements

Significant changes were observed in all measured CoP variables during stair de-scent among EG participants following the 10-week intervention (

Table 4. CoP parameters during stair descent (mean ± SD).

		Group
Variables	Training	EG (N = 15)	CG (N = 15)
AP Range (cm)	pre	29.73 ± 1.68	31.10 ± 12.38
	post	37.85 ± 12.24 *	30.64 ± 16.56
AP Excursion (cm)	pre	38.34 ± 3.53	40.46 ± 15.57
	post	58.15 ± 12.85 ***, †	42.40 ± 17.34
AP Velocity (cm/s)	pre	6.53 ± 0.73	6.89 ± 2.35
	post	9.74 ± 1.88 ***, †	7.12 ± 2.34
AP RMS (cm)	pre	6.68 ± 0.67	6.36 ± 1.29
	post	1.48 ± 0.81 ***, †	6.42 ± 1.42
ML Range (cm)	pre	24.78 ± 1.95	19.59 ± 9.57
	post	12.54 ± 7.06 ***, †	19.51 ± 7.87
ML Excursion (cm)	pre	28.56 ± 2.11	26.89 ± 10.16
	post	19.38 ± 7.15 ***, †	28.12 ± 10.43
ML Velocity (cm/s)	pre	4.11 ± 0.85	3.65 ± 1.44
	post	2.88 ± 0.99 **, †	3.80 ± 1.40
ML RMS (cm)	pre	7.44 ± 0.69	7.06 ± 1.38
	post	4.51 ± 0.78 ***, †	6.62 ± 0.97
Total Excursion (cm)	pre	46.84 ± 3.32	52.95 ± 10.86
	post	54.86 ± 13.57 *	58.47 ± 14.58
Total Velocity (cm/s)	pre	6.98 ± 1.09	7.60 ± 2.09
	post	9.54 ± 1.60 ***, †	7.89 ± 1.96

AP, antero-posterior; ML, mediolateral; RMS, root mean square. * p < 0.05, ** p < 0.01, *** p < 0.001: pre- vs. post-training, † p < 0.05, EG vs. CG groups. All parameters differed significantly.

). In the AP direction, the EG exhibited significant increases in AP range, excursion, and velocity, accompanied by a marked reduction in AP RMS. Additionally, significant decreases were observed in ML range, excursion, velocity, and RMS. The EG also demonstrated significant increases in total excursion and total velocity.

Furthermore, significant differences between the EG and CG were identified in AP excursion, AP velocity, AP RMS, ML range, ML excursion, ML velocity, ML RMS, and total velocity. In contrast, no significant changes were observed within the CG.

3.4. Joint Kinematics

After completing the 10-week Aqua Vest unstable exercise program, participants in the EG group exhibited significant changes in many lower extremity kinematic variables during stair ascent and descent (

Table 5. Angle ROM parameters during stair ascent and descent (mean ± SD).

			Stair Ascent		Stair Descent
Angle Variables (°)		Training	EG	CG	EG	CG
Ankle Dorsi–plantar flexion ROM	Non-dominant	pre	39.46 ± 5.96	44.29 ± 6.86	28.71 ± 6.26	27.83 ± 9.66
		post	42.77 ± 4.33 *	45.38 ± 7.23	34.40 ± 10.36 *	29.79 ± 8.49
	Dominant	pre	39.15 ± 6.83	43.53 ± 6.22	29.89 ± 4.35	28.52 ± 7.95
		post	41.72 ± 5.59	45.13 ± 5.92	33.81 ± 11.34	31.28 ± 8.64
Ankle Inversion–eversion ROM	Non-dominant	pre	4.65 ± 1.81	3.60 ± 1.68	2.46 ± 1.08	2.73 ± 1.43
		post	4.15 ± 1.99	3.12 ± 1.61	2.73 ± 1.55	3.03 ± 1.69
	Dominant	pre	4.03 ± 2.33	3.13 ± 1.86	2.41 ± 2.02	3.39 ± 3.03
		post	3.16 ± 1.53	2.96 ± 1.96	2.05 ± 1.52	5.46 ± 9.46
Hip Flexion–extensionROM	Non-dominant	pre	57.53 ± 2.84	58.11 ± 3.97	9.67 ± 3.69	10.75 ± 3.15
		post	52.09 ± 4.04 *, †	57.22 ± 3.29	6.75 ± 3.85 *, †	11.74 ± 5.22
	Dominant	pre	58.17 ± 3.71	57.42 ± 3.45	9.86 ± 3.35	10.75 ± 5.48
		post	50.99 ± 3.49 *, †	56.76 ± 4.45	7.42 ± 3.59 *, †	10.55 ± 4.00
Hip Adduction–abduction ROM	Non-dominant	pre	14.23 ± 3.79	13.07 ± 6.01	6.87 ± 2.60	7.24 ± 3.92
		post	13.51 ± 3.57	13.40 ± 5.96	7.49 ± 3.53	6.64 ± 3.62
	Dominant	pre	15.04 ± 3.58	13.10 ± 4.12	6.29 ± 2.41	7.11 ± 3.70
		post	14.62 ± 4.69	13.64 ± 4.89	7.21 ± 3.39	7.77 ± 4.15
Knee Flexion–extension ROM	Non-dominant	pre	56.22 ± 3.55	56.95 ± 6.95	71.79 ± 4.90	72.15 ± 4.63
		post	46.84 ± 4.18 *, †	53.68 ± 4.58	77.09 ± 7.19 *, †	70.39 ± 5.95
	Dominant	pre	56.13 ± 2.82	54.66 ± 4.77	69.52 ± 6.32	72.37 ± 6.01
		post	46.49 ± 4.97 *, †	53.88 ± 3.93	69.09 ± 9.42†	74.65 ± 4.53
Knee Adduction–abduction ROM	Non-dominant	pre	18.86 ± 5.20	15.76 ± 8.03	8.75 ± 6.40	13.57 ± 8.68
		post	17.99 ± 3.74	18.44 ± 11.42	10.02 ± 5.79	15.87 ± 10.27
	Dominant	pre	13.50 ± 11.16	10.45 ± 7.13	15.53 ± 15.06	19.90 ± 12.89
		post	9.72 ± 8.40	9.91 ± 4.23	13.11 ± 12.42	18.62 ± 13.20

ROM, range of motion. * p < 0.05: pre- vs. post-training, † p < 0.05, EG vs. CG groups. All parameters differed significantly.

). No differences in any of these variables were observed between groups at baseline. When assessing the ankle dorsiflexion–plantar flexion ROM while ascending and descending stairs, a significant post-training increase was observed for the left ankle of participants in the EG group, consistent with the enhancement of flexibility relative to individuals in the CG. With respect to hip kinematics, EG group participants exhibited notable post-training reductions in flexion–extension ROM relative to the CG, particularly when descending stairs. Analyses of knee kinematics also revealed significant post-training differences in knee flexion–extension ROM. There were also significant decreases in knee flexion–extension ROM for both the left and right legs for EG participants during stair ascent, together with a significant increase in left-sided ROM during stair descent. However, no significant differences in ankle inversion–eversion, hip adduction–abduction, or knee adduction–abduction ROM within or between groups after the exercise intervention.

4. Discussion

This study aimed to assess the effects of a 10-week Aqua Vest-based unstable load training program on lower extremity kinematics and pain reduction in middle-aged women with degenerative knee arthritis (DKA) during stair ascent and descent. The findings revealed significant improvements in postural stability, gait performance, and pain reduction among participants in the exercise group (EG) compared to those in the control group (CG).

Stair ambulation imposes greater biomechanical demands than level walking, re-quiring increased hip, ankle, and knee joint flexion angles [40] and generating higher knee joint loads [41]. Prior research by Son and Kim [42] indicated that DKA patients experience significant changes in joint angles during stair ascent, which can exacerbate symptoms due to increased knee joint loading. In the present study, after the 10-week intervention, EG participants exhibited a significant reduction in hip flexion angles during both stair ascent and descent, indicating enhanced hip joint stability. Additionally, knee joint flexion angles significantly decreased during stair ascent, suggesting improved knee joint stability, while knee flexion–extension ROM significantly increased during stair descent, reflecting enhanced knee joint flexibility. These contrasting adaptations highlight the differential biomechanical effects of unstable load training on stair ascent versus descent, warranting further investigation into the underlying mechanisms.

Gait variability and reduced complexity are commonly associated with impaired mobility and increased fall risk [43]. DKA patients often exhibit altered gait patterns, including reduced stride length and walking speed, as part of a more cautious gait strategy [44]. In this study, EG participants demonstrated significant post-intervention improvements, including increased walking speed and reduced stride and step times during stair ambulation. These results suggest that the Aqua Vest-based unstable exercise program effectively enhanced stair gait performance while concurrently reducing knee pain, underscoring its potential for inclusion in rehabilitative and clinical interventions for DKA patients.

Mediolateral (ML) sway is a critical factor in maintaining balance and is closely associated with fall risk. ML CoP parameters (distance and velocity) are frequently used as indicators of postural control deficits and fall susceptibility [12,45]. After the intervention, EG participants exhibited significant reductions in ML CoP parameters, suggesting enhanced lateral postural control and stability. This improvement was likely due to the exercise regimen’s emphasis on trunk control and activation of core and lower extremity muscles, facilitating the adoption of a more stable lateral balance strategy and reducing dynamic postural fluctuations.

The effectiveness of resistance training using unstable surfaces and devices for rehabilitation has been widely established. Unstable loads increase time under tension, potentially leading to distinct training adaptations. In summary, unstable load training serves as an effective supplemental exercise by enhancing time under tension and maintaining high muscle activation across both primary and stabilizing muscles, even with lighter loads [35].

Performing exercises on unstable surfaces, such as the BOSU, increases co-contraction and antagonist muscle activity. While enhanced antagonist activity contributes to joint and muscle protection, it also improves motor control and balance [32]. According to Kibele and Behm [46], unstable loads or bases can generate high-intensity internal muscle contractions despite reduced external force output, creating an effective training stimulus compared to stable conditions. Under unstable conditions, even a relatively small resistive torque at the distal limb can produce substantial motive torque through the torso.

Nairn et al. [47] investigated muscle activity during squats using three instability tools (bar, BOSU, and water tube) and found that different instability sources altered electromyographic and kinematic patterns. These variations have implications for technique learning, rehabilitation, and training goal setting. Their findings suggest that selecting the appropriate instability tool should be guided by specific rehabilitation objectives.

However, not all unstable devices are equally effective. Wahl and Behm [48] found that certain unstable surfaces failed to significantly increase muscle activity during squats, particularly in highly resistance-trained individuals. Moderately unstable devices, such as the BOSU, did not enhance muscle activation in this population. Consequently, most research has focused on comparing the short-term effects of different instability tools. Among these, water-based instability devices have gained increasing attention for their applications in both performance enhancement and rehabilitation.

Studies suggest that training with water-filled devices enhances muscle activation and challenges postural stability more effectively than stable load exercises [1,33,49,50]. Ditriolo et al. [1] demonstrated that exercises with a water-filled tube (WT) resulted in greater core muscle activation and posed a more significant challenge to postural stability compared to traditional barbell training. Similarly, Calatayud et al. [29] reported that performing clean and jerk lifts with a water bag elicited a stronger core training stimulus than using barbells or sandbags of equivalent weight. Nairn et al. [49] further analyzed the effects of WT induced instability during the bench press, showing increased activation of the oblique and erector spinae muscles.

Nevertheless, previous research has suggested that the primary contribution of instability training may be the improvement of core stability rather than limb strength. Furthermore, the initial goal of instability training may not necessarily be significant strength gains but rather an enhancement of balance, stability, and proprioceptive capabilities [30]. Water inertia-based unstable exercises also have beneficial effects on postural stability, inducing greater medial–lateral movement. Nairn et al. [49] further noted that the introduction of the WT significantly affected the mediolateral direction, with the impact being more pronounced in the range of the center of pressure (CoP) than in its sway velocity. This effect is attributed to the shape and size of the WT, which continuously alters the weight distribution as the water flows along its main axis.

The Aqua Vest introduces an innovative approach to unstable load training by incorporating partially filled water tubes that generate dynamic weight distribution shifts. Furthermore, Calatayud et al. [29] suggested that the fluidity of the water in a water bag could introduce greater asymmetrical disturbances during exercises, especially in the final phases of movement. These findings suggest that tools utilizing water inertia may offer advantages in sports that demand enhanced somatosensory or proprioceptive contributions to balance control. The inertia of the water responds dynamically to body movements, continuously altering the load distribution and creating an inherently unstable resistance. This instability enhances neuromuscular control, postural adjustments, and core muscle activation. Glass and Albert [50] demonstrated that adjusting water volume in such devices modulated inertial perturbation, offering a scalable challenge for users. The dynamic variability induced by water inertia aligns with principles of dynamic systems theory and ecological movement approaches, fostering adaptability in movement patterns [36].

A key advantage of the Aqua Vest over other unstable load devices is its ability to leave the arms and hands free, allowing for greater movement flexibility and coordination during training. Unlike traditional unstable load tools, which may limit upper limb mobility, the Aqua Vest facilitates whole-body dynamic balance. Despite its potential, the effects of unstable training using wearable water-filled devices on postural stability and gait function remain underexplored. This study addresses this gap by investigating the influence of Aqua Vest on gait and postural stability in middle-aged women with DKA. This highlights the novelty of the present study and the need for further investigations to explore its full rehabilitative potential. Meanwhile, the portability of Aqua Vest presents a practical option for home-based rehabilitation, an area that has not been widely studied in previous research.

This study has several limitations. Firstly, this study focused on lower extremity kinematics and postural stability but did not assess joint kinetics, muscle activation, or load distribution. The findings confirm that water inertia-based training improves functional performance and reduces pain in individuals with degenerative knee arthritis (DKA), laying the groundwork for future research on its effects on joint moments and ground reaction forces (GRF). Notably, it did not compare different phases of the gait cycle, limiting the analysis of movement changes. The small changes in knee and hip adduction–abduction range of motion (ROM) suggest that the exercise protocol may need adjustments to improve frontal plane stability. Future studies should explore modifications to training methods that enhance muscle engagement in this plane for better results.

Secondly, the lack of electromyography (EMG) analysis limits insight into muscle contributions to postural stability. Future studies should incorporate EMG and kinetic assessments to better understand neuromuscular adaptations. Advanced biomechanical modeling tools, such as OpenSim musculoskeletal model or AI-driven algorithms, could provide more precise estimates of knee contact force changes and muscle forces [51]. Moreover, integrating recurrence quantification analysis, EEMD-DFA algorithms, or vibroarthrography may improve the accuracy of knee joint function and cartilage health assessments, leading to better data visualization and greater clinical applicability [52,53,54].

Furthermore, the sample size (N = 30) was small, which may affect the reliability and applicability of the results. In addition, the study was limited to middle-aged women with degenerative knee arthritis, restricting the generalizability of the findings to other populations. Differences in sex, age, and health conditions may influence the effects of water inertia-based unstable exercise, and it remains unclear whether similar benefits can be observed in younger individuals, men, or those with different musculoskeletal conditions. To establish more effective exercise therapy strategies for different populations, it is necessary to compare Aqua Vest with other water inertia-based training tools in controlled experiments. Expanding research to include more diverse populations and conducting statistical power analysis will further improve accuracy and confirm the broader applicability of these training methods.

Finally, the study lasted 10 weeks but did not measure how long the effects lasted after the program ended. Without follow-up tests, it is unclear if the improvements in movement and balance remain over time. The lasting effects of water inertia-based unstable exercise may differ depending on a person’s activity level, muscle strength, and daily habits. Follow-up assessments at different times are needed to understand how long the benefits last. This will also help develop more tailored exercise therapy strategies suited to different individuals and optimize the application of water inertia-based unstable exercise in various clinical and rehabilitation settings.

Overall, this study demonstrated that a 10-week Aqua Vest-based unstable load exercise program led to significant improvements in gait performance, postural stability, and knee pain reduction in middle-aged women with DKA. These findings support the hypothesis that unstable exercises leveraging water inertia positively influence postural control, lower extremity function, and pain during stair ambulation. The Aqua Vest represents a promising tool for rehabilitation and fall prevention in individuals with DKA, warranting further research to refine its application in clinical and sports performance settings.

5. Conclusions

The Aqua Vest-based unstable load training proved to be an effective intervention for improving functional outcomes and reducing pain in middle-aged women with degenerative knee arthritis. By leveraging the inertial properties of water, this training method introduced controlled instability, enhancing balance, proprioception, and neuromuscular adaptations critical for postural control. These findings underscore its potential as a rehabilitation strategy for mitigating fall risk, improving dynamic balance, and alleviating pain during stair ambulation. Additionally, it contributed to enhanced postural stability, particularly in the mediolateral direction during stair descent. Future research should further explore the therapeutic potential of water-based unstable load training across diverse patient populations to refine its application in rehabilitation and clinical practice.