The effects of aquatic therapy combined with transcranial direct current stimulation (tDCS) on proprioception and gait speed in older adults with knee osteoarthritis: an eight-week randomized sham-controlled trial

Abstract

Background

Knee osteoarthritis (KOA) is a prevalent condition among older adults, leading to impaired proprioception and reduced gait speed, which compromise mobility and quality of life. While aquatic therapy and transcranial direct current stimulation (tDCS) have individually shown potential for enhancing motor and sensory functions, their combined effects are not well understood.

Objective

This study aimed to evaluate the combined effects of aquatic therapy and tDCS on knee proprioception and gait speed in older women with KOA.

Methods

A randomized, triple-blind, parallel-group, sham-controlled clinical trial was conducted with 68 elderly women with KOA (aged 63–68 years). Participants were allocated into four groups: (1) Aquatic Therapy + Sham tDCS (2), Aquatic Therapy + Real tDCS (3), Real tDCS, and (4) Sham tDCS. Interventions were conducted twice weekly over an 8-week period. Knee proprioception was assessed using a digital goniometer, and gait speed was measured via the 10-Meter Walk Test under normal and dual-task conditions. Data were analyzed using analysis of covariance (ANCOVA), controlling for height and baseline scores.

Results

Paired t-tests revealed significant improvements in gait speed (under both normal and dual-task conditions) and knee proprioception in all intervention groups, except the Sham tDCS group (p ≤ 0.05). Analysis of covariance (ANCOVA), controlling for height and pre-test scores, showed significant between-group differences in both gait speed and knee proprioception. Post hoc comparisons indicated that the Aquatic Therapy + Real tDCS group exhibited significantly greater improvements in knee proprioception compared to both the Aquatic Therapy + Sham tDCS and Real tDCS groups (p ≤ 0.05), supporting the additive effect of real tDCS on sensory outcomes. However, no significant differences in gait speed were observed between the Aquatic Therapy + Real tDCS and Aquatic Therapy + Sham tDCS groups under either normal or dual-task conditions.

Conclusion

The combined use of aquatic therapy and tDCS appears to enhance proprioceptive function more effectively than either intervention alone, while improvements in gait speed seem primarily driven by aquatic therapy. These findings suggest that tDCS may offer sensory-specific benefits in multimodal rehabilitation for older adults with KOA. Further research is warranted to assess the long-term effects, evaluate outcomes under varied task demands, and examine the generalizability to broader populations.

Trial registration

This clinical trial was registered with Iran’s Clinical Trial Registration Center (IRCT)[IRCT20190908044722N8] on 8 January 2025.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12877-025-06253-5.

Introduction

Knee osteoarthritis (KOA) is a common and growing health issue worldwide, especially among older adults. As of 2019, about 364.6 million people were living with KOA, with women affected more often than men [1]. Studies show that more than half of women and over a third of men above age 50 experience knee pain linked to this condition [2]. The condition is characterized by progressive cartilage loss, leading to pain, stiffness, and functional impairment, and is strongly associated with aging, obesity, joint injury, and genetic predispositions [3]. Among individuals over 60, KOA affects approximately 13% of women and 10% of men, significantly reducing mobility and quality of life [3–5].

Pain and mobility impairment components of KOA pose significant challenges to physical functioning, resulting in disability for approximately 66% of affected individuals and accounting for 2.4% of all global years lived with disability [6]. These statistics underscore the urgent need for effective interventions that reduce symptoms and restore function. Two critical determinants of functional independence as well as quality of life in patients with KOA are proprioception and gait speed. Gait speed, sometimes called the sixth vital sign, is a well-established predictor of disability, falls risk, and death [7–9], and is favored for its simplicity, cost-effectiveness, and strong association with diverse health outcomes [10, 11]. Similarly, proprioception is critical to knee joint stability, balance, and integrated motion. In KOA, degradation of the mechanoreceptors blunts proprioceptive feedback, producing joint instability, enhanced risk of falls, and greater disability [12, 13]. As such, both gait speed and proprioception are commonly employed as reliable and sensitive indicators of the success of therapies in clinical trials of KOA [14, 15].

Conservative management of KOA, including standard rehabilitation and exercise programs, is largely meant to ease pain and improve muscular strength. While these therapies may reduce symptoms and strengthen the muscles, their effects on proprioceptive deficits and gait disturbances remain unclear [16, 17]. While certain protocols have shown minor improvements in the proprioception of the affected joint, these are commonly not enough to restore normal sensorial function in KOA patients [18, 19]. Similarly, findings related to gait speed and coordination remain inconsistent, with the majority of studies recording small or non-significant improvements [20, 21]. These limitations highlight the need for adjunctive or alternative therapies that specifically target the sensorimotor impairments of mobility and balance. Recent systematic and comparative reviews noted the potential of new therapeutic techniques, specifically transcranial direct current stimulation (tDCS) and aquatic therapy, to improve proprioception and gait performance in patients with KOA [18, 22].

Innovative interventions such as aquatic therapy and transcranial direct current stimulation (tDCS) may address limitations of conventional rehabilitation approaches. Aquatic therapy utilizes the specific physical characteristics of the aquatic environment, buoyancy, resistance, and hydrostatic pressure to enhance lower limb strength, proprioception, and gait speed [23–27]. For instance, an eight-week aquatic training regimen improved proprioception and fast-paced gait speed on the 40-meter fast-paced walk test significantly [24]. In contrast, tDCS exerts its effects via neuromodulation, inducing cortical plasticity that can facilitate motor control and enhance proprioceptive accuracy [28]. A body of literature increasingly supports the effectiveness of tDCS to modulate gait and balance in various clinical populations. Toktas et al. and Yang et al. for instance, demonstrated significant gains in gait function and balance in post-stroke individuals, ultimately contributing to enhanced mobility and quality of life [29, 30]. Additionally, a recent meta-analysis by Wu et al. found that while tDCS effectively reduced pain in the short term among KOA patients, its long-term effects on pain relief and physical function remained limited [31].

Even though both tDCS and aquatic therapy individually proved beneficial, Monomodal interventions may be insufficient to entirely overcome proprioceptive and gait defects in elders with KOA. For instance, Xiao et al. reported inconsistent improvements in proprioception and balance following a 24-week tDCS intervention [32]. Conversely, tDCS-assisted physical therapies were found to be beneficial by studies such as Rahimi et al. and Teixeira et al. yielding significant improvements in pain alleviation and functional capacity in patients with KOA [33, 34]. Although promising, very little is known about the combined effect of these therapies in the elderly population. Mechanistically, tDCS modulates excitability of the cortex through polarity-selective alterations in the membrane potential and efficacy of the synapse, largely mediated through the N-methyl-D-aspartate (NMDA) receptor networks. Anodal tDCS enhances the excitability of the cortex and facilitates the neuroplasticity of the type involved in learning of the limb movements and rehabilitation [35, 36]. At the same time, aquatic therapy facilitates gains in muscle strength, joint stability, and proprioception, as the properties of the water itself provide an environment conducive to these effects, including resistance and buoyancy. This is particularly beneficial in KOA, where proprioception impairment makes the joint and the body as a whole much more prone to instability and imbalance and the consequent risk of falls [37, 38]. Recent studies suggest that the combination of tDCS, acting on the central mechanism, and aquatic therapy, acting on the peripheral mechanism, may provide synergistic gains in proprioception and gait function [39, 40]. This integrative approach leverages both neuroplastic and biomechanical mechanisms, potentially providing a more effective avenue for functional recovery in this population [39].

This trial aims to explore whether combining aquatic therapy with transcranial direct current stimulation (tDCS) can more effectively improve proprioception and gait speed in older adults with knee osteoarthritis. By addressing both the neural and physical aspects of movement, we hope to go beyond the limitations of traditional rehabilitation. This study aims not only to improve proprioception and gait speed but also to enhance overall functional mobility and independence among older adults with KOA, through a multimodal, neurophysiological informed intervention.

Methodology

Study Design

The study employed a randomized, triple-blind, sham-controlled clinical trial design, ensuring blinding of the therapist, evaluator, and participant. The Ethics Committee of Urmia University approved the study (ID IR.URMIA.REC.1402.026). This study was registered as a clinical trial on the Iranian Registry of Clinical Trials (https://irct.behdasht.gov.ir/) (IRCT20190908044722N8) on 8 January 2025. Although all study procedures were implemented prospectively, trial registration with the Iranian Registry of Clinical Trials (IRCT ID: IRCT20190908044722N8) was completed retrospectively on January 8, 2025 (https://irct.behdasht.gov.ir/) due to administrative delays unrelated to the study design or execution. The authors acknowledge that prospective registration is the preferred standard as per ICMJE and CONSORT recommendations and affirm that all aspects of trial conduct, including randomization, allocation concealment, blinding, and outcome assessment were performed with strict adherence to ethical and methodological principles. Written informed consent was obtained from all participants prior to enrollment, with explicit assurance of their right to withdraw at any time without consequences. The overall study flow is summarized in Fig. 1.

Fig. 1.

Flow diagram of the study

Participants

A total of 68 elderly women with knee osteoarthritis (KOA) were recruited through a non-random, convenience sampling method. Eligibility criteria included women aged 63–68 years, diagnosed with KOA by a certified rheumatologist using X-ray imaging and physical examination, and inactive individuals performing less than one hour of physical activity per week [40]. All participants completed the Kellgren-Lawrence (KL) grading scale within the preceding six months, with a KL grade of ≥ 1 serving as an inclusion criterion [41]. Participants also needed to be hemodynamically stable (blood pressure < 140 and 90 mmHg) [42], with a body mass index (BMI) < 40 kg/m², and without severe medical conditions, brain implants, or cardiac pacemakers. Exclusion criteria included a history of knee surgeries or other joint diseases affecting gait, Untreated thyroid dysfunction (hypothyroidism or hyperthyroidism), Severe respiratory, neurological, or musculoskeletal disorders, contraindications to transcranial direct current stimulation (tDCS), such as epilepsy, brain tumors, or skin lesions near electrode application sites [42], cortisone injection in their knee joint within the past month [43], left ventricular dysfunction, symptoms of cold, infections, or other illnesses during the study period.

Sample size calculation

The sample size was determined using G*Power 3.1.9.7, selecting the ANCOVA model (fixed effects, omnibus, one-way) to align with the study design, which included four intervention groups and two covariates. Based on a moderate effect size (f = 0.25), a significance level (α) of 0.05, and a statistical power (1–β) of 0.90, the minimum required sample size was determined to be 61 participants. To account for potential attrition, an estimated dropout rate of 10% was considered. Therefore, the target sample was increased to 68 participants to ensure adequate power in the event of participant withdrawal.

Randomization

The sequence of random grouping was generated using Random Allocation Software version 1.0.0 (freeware for Windows XP/Vista/7/8/10/11). Elderly female participants (n = 68) who met the inclusion criteria were randomly assigned in a 1:1:1:1 ratio to one of four groups: Aquatic therapy + Sham tDCS, Aquatic therapy + Real tDCS, Real tDCS, or Sham tDCS. To ensure the integrity of the randomization process, a single individual, blinded to group allocation, handled participant enrollment and group assignments.

Blinding

This study employed a randomized, triple-blind, sham-controlled design to minimize bias. Blinding was maintained at three levels: participants, outcome assessors, and the therapist administering the intervention. Participants were unaware of whether they received active or sham tDCS, as the device’s double-blind mode automatically delivered either condition based on a pre-programmed, password-protected protocol. In the sham condition, the device simulated the initial tingling sensation associated with active tDCS, thereby maintaining participant blinding. The therapist, responsible solely for electrode placement and session initiation, had no access to information regarding stimulation allocation. To reinforce therapist blinding, an external researcher not involved in the intervention or outcome assessment pre-programmed the device. Outcome assessments were conducted by an independent evaluator blinded to group assignments. Finally, statistical analyses were performed separately by a researcher who was aware of group allocations to ensure correct interpretation of the data while maintaining the integrity of the blinding process.

Data collection instruments

Primary outcomes

The study included two co-primary outcomes:

Knee proprioception

In order to evaluate and test the sense of the knee joint position, a digital goniometer with an accuracy of 0.1 degrees was used. It was made by Zanij Industrial Research and Development Company of Iran. The device includes a digital protractor system, reader, calibration module, data recording, and processing software and two fixed and movable arms are made of aluminum. Participants were seated with their knee flexed at 85°, and they were asked to replicate a target knee angle of 60° without visual feedback. Three trials were conducted, and the average error from the target angle was recorded as the measure of knee proprioception [44]. The right knee underwent a proprioceptive evaluation.

Gait speed

The gait speed was evaluated by using the 10-Meter Walk Test under two conditions:

Gait: Subjects were requested to walk at a comfortable speed over the 10 m distance. Time was recorded for two trials, and the average used for analysis.

Cognitive Task Condition: In this condition, participants walked the same 10-meter distance while concurrently performing a visuospatial N-back task. This task required them to respond to visual stimuli by saying “similar” or “dissimilar.” The average time of two trials was recorded for analysis [45].

Procedure

Participants were recruited from a clinic in Ardabil city by reviewing patient records from the past two years. Eligible individuals were invited to a meeting where the study objectives, procedures, and potential side effects were thoroughly explained. A rheumatologist confirmed the diagnosis of knee osteoarthritis (KOA) based on clinical evaluation and X-ray imaging. Only participants with a Kellgren-Lawrence (KL) grade of ≥ 1, assessed within the preceding six months, were included [41].

Following informed consent and baseline (pre-test) assessments, 68 eligible elderly women were randomly assigned, in a 1:1:1:1 ratio, to one of four groups based on Kaski et al.’s design [46]: [1] Aquatic Therapy + Sham tDCS [2], Aquatic Therapy + Real tDCS [3], Real tDCS only, or [4] Sham tDCS only. Randomization was performed by an independent researcher using a computer-generated sequence. All evaluations were conducted by an assessor blinded to group allocations. Participant recruitment and assessments took place at a clinic in., while interventions were conducted at.Jam Swimming Pool under the supervision of an experienced exercise professional and a medical doctor. Participants were permitted to continue prescribed medications initiated at least 30 days before enrollment.

Aquatic Therapy Intervention.

Aquatic therapy sessions were held twice weekly for 8 weeks. Each 60-minute session included a 15-minute warm-up, 45 min of functional and proprioceptive exercises (e.g., walking, side-stepping, mini squats, hip movements, arm exercises), and a 5-minute cool-down [47]. Exercises were performed in chest-deep water (1.2–1.3 m depth) at a water temperature of 34–36 °C and an air temperature of 24 °C. Water resistance was naturally generated by movement through the water. Exercise intensity was monitored using the Borg Rating of Perceived Exertion (RPE) scale, maintaining moderate intensity (RPE 12–14) [48]. Exercise volume progressed from three repetitions per exercise at the start to three sets of ten repetitions by the end of the program. The duration and frequency of the program were based on Krishnan et al. [49]. A lifeguard was present during all sessions, and a physician supervised the intervention. Sham tDCS was administered without active current.

tDCS Intervention.

Real tDCS: Participants received anodal stimulation (2 mA for 30 min) using the Neurostim2 two-channel device, targeting the primary motor cortex (M1, C3) and the right posterior parietal cortex (PPC, P4), with cathodes placed over Fp1 and Fp2. A 10-second ramp-up and ramp-down were applied to minimize discomfort. This dual-site protocol was informed by previous studies suggesting that anodal tDCS over M1 can enhance gait speed [50], while stimulation of the PPC is associated with improvements in proprioception [51].

Sham tDCS: Participants received sham stimulation mimicking the initial sensation without active current.

Participants in the Aquatic Therapy + Real tDCS group received Real tDCS before aquatic sessions; those in the Aquatic Therapy + Sham tDCS group received sham stimulation. Participants in the Real tDCS and Sham tDCS groups (without aquatic therapy) followed an identical stimulation protocol.

Analysis of data

All statistical analyzes were performed with SPSS^® 26.0. The nominal level of significance was set at alpha < 0.05. Shapiro-Wilk test, Levin’s F test and ANCOVA test were used to check the normality of data distribution, homogeneity of variances and homogeneity of regression slope. Paired t-test was used to examine the intragroup effects of gait speed and Knee proprioception. To compare the level of gait speed and Knee proprioception between the research groups, the covariance test was used along with Bonferroni’s post hoc test, and in this test, the pre-test of the research variables was considered as covariate.

Results

Patient characteristics

The characteristics of the participants are presented in Table 1. One-way ANOVA was used to compare continuous variables (age, weight, height, and disease duration) across the four groups. For categorical variables (radiological stage), chi-square tests were conducted.

Table 1.

Demographic and clinical characteristics of participants across groups

Characteristics	Aquatic Therapy + Sham tDCS (n = 17)	Real tDCS (n = 17)	Aquatic Therapy + Real tDCS (n = 17)	Sham tDCS (n = 17)
Age (years)	65.05 ± 1.34	65.00 ± 1.50	65.47 ± 1.62	65.17 ± 1.66
Weight (kg)	81.29 ± 7.68	75.00 ± 18.61	79.58 ± 6.90	76.94 ± 7.53
Height (m)	1.636 ± 0.13	1.586 ± 0.020	1.730 ± 0.068	1.618 ± 0.027
Mean disease duration (years)	3.41 ± 1.41	3.11 ± 0.92	2.70 ± 1.15	3.70 ± 1.53
Radiological stage (K٫L) Left
Stage 2	7(41.2%)	5(29.4%)	8(47.1%)	7(41.2%)
Stage 3	10(58.8%)	12(70.6%)	9(52.9%)	10(58.8%)
Right
Stage 2	10(58.8%)	6(35.3%)	8(47.1%)	7(41.2%)
Stage 3	7(41.2%)	11(64.7%)	9(52.9%)	10(58.8%)

Data are presented as mean ± SD; tDCS: direct transcranial electrical stimulation; kg kilograms, m meters

The four groups did not differ significantly in age (F (3, 64) = 0.31, p = 0.81) or weight (F (3, 64) = 1.03, p = 0.38) or mean disease duration (F (3, 64) = 1.88, p = 0.14). However, a significant difference was observed in height among the groups (F (3, 64) = 10.70, p = 0.001). Chi-square tests revealed no significant differences in radiological stage (Kellgren-Lawrence grading) among the four groups for both right (χ² (3) = 2.07, p = 0.55) and left knees (χ² (3) = 2.14, p = 0.49).

The results of the paired t-test analysis are summarized in Table 2, highlighting the effects of aquatic therapy and tDCS on gait speed under normal conditions, gait speed in dual-task conditions, and knee proprioception in female older adults with knee osteoarthritis.

Table 2.

Differences in proprioception of knee and gait speed among female older adults with knee osteoarthritis at baseline and post-intervention

	Aquatic Therapy + Sham tDCS		Within-group comparison		Real tDCS (n = 17)		Within-group comparison		Aquatic Therapy + Real tDCS (n = 17)		Within-group comparison		Sham tDCS (n = 17)		Within-group comparison
	Pre- test	Post- test	t	p	Pre- Test	Post- Test	t	p	Pre- test	Post- test	t	p	Pre- test	Post- test	t	p
Gait Speed under normal conditions (second)	12.18 ± 2.18	10.10 ± 1.42	4.97	0.001*	12.75 ± 1.81	11.35 ± 1.33	2.379	0.030*	12.32 ± 1.03	9.87 ± 1.38	7.111	0.001*	12.11 ± 1.54	12.14 ± 1.56	− 0.612	0.549
Gait Speed in Dual Task(scond)	14.68 ± 2.51	12.52 ± 2.12	3.317	0.004*	15.01 ± 2.02	13.71 ± 1.76	2.827	0.012*	14.41 ± 1.84	12.09 ± 1.89	4.151	0.001*	14.55 ± 1.58	14.66 ± 1.23	− 0.320	0.753
proprioception of knee(degree)	11.97 ± 3.62	11.41 ± 3.53	5.546	0.001*	12.88 ± 4.23	12.60 ± 4.31	8.004	0.001*	17.46 ± 6.40	14.46 ± 6.40	4.74	0.001*	14.59 ± 3.29	14.41 ± 3.18	2.012	0.061

*: P ≤ 0.05

Paired t-tests were conducted to examine the within-group effects of the interventions. Significant improvements were observed in several parameters among female older adults with knee osteoarthritis, as shown in Table 2.

For gait speed under normal conditions, significant improvements were observed in the Aquatic Therapy + Sham tDCS group, the Real tDCS group, and the Aquatic Therapy + Real tDCS group. However, the Sham tDCS group did not exhibit significant changes in gait speed under normal conditions. Cohen’s d values indicated a very large effect in the Aquatic Therapy + Real tDCS group (d = 2.01) and the Aquatic Therapy + Sham tDCS group (d = 1.13), a large effect in the Real tDCS group (d = 0.88), and a negligible effect in the Sham tDCS group (d = −0.02).

For gait speed in dual task conditions, significant improvements were noted in the Aquatic Therapy + Sham tDCS group, the Real tDCS group, and the Aquatic Therapy + Real tDCS group. The sham tDCS group again showed no significant changes. Cohen’s d values revealed a very large effect in the Aquatic Therapy + Real tDCS group (d = 1.24), a large effect in the Aquatic Therapy + Sham tDCS group (d = 0.93), a medium to large effect in the Real tDCS group (d = 0.68), and a very small effect in the sham tDCS group (d = 0.07).

For proprioception of the knee, statistically significant improvements were observed in the Aquatic Therapy + Sham tDCS group, the Real tDCS group, and the Aquatic Therapy + Real tDCS group. However, the Sham tDCS group did not show significant improvements. Notably, although the Real tDCS group showed a statistically significant improvement, the effect size was very small (Cohen’s d = 0.07), indicating a limited practical or clinical impact. Cohen’s d values indicated a medium effect in the Aquatic Therapy + Real tDCS group (d = 0.47), a small effect in the Aquatic Therapy + Sham tDCS group (d = 0.16), and very small effects in both the Real tDCS (d = 0.07) and Sham tDCS (d = 0.06) groups. Table 3

Table 3.

Effects of aquatic therapy and tDCS on proprioception of knee and gait speed among female older adults with knee osteoarthritis

Dependent Variable	Type III Sum of Squares	Degrees of freedom	Mean Square	F	Significance	Partial Eta Squared
Gait Speed under normal conditions	49.71	3	16.57	9.86	0.001*	0.323
Gait speed in Dual Task	42.047	3	14.016	5.37	0.002*	0.206
proprioception of knee	56.990	3	18.997	218.541	0.001*	0.914

*Statistically significant at p ≤ 0.05;

Analysis of covariance (ANCOVA) was performed to assess between-group differences while controlling for height, which varied significantly across the groups. Height was included as a covariate due to its significant negative correlations with gait speed under dual-task conditions (r = − 0.343, p = 0.004) and normal conditions (r = − 0.263, p = 0.030). Although no significant correlation was found between height and knee proprioception (r = 0.134, p = 0.275), height was retained as a covariate in all ANCOVA models for consistency and to control for potential confounding.

The results of the ANCOVA revealed that, after controlling for the effects of both pre-test scores and height as covariates, both gait speed and proprioception of the knee were significantly affected by the interventions (Aquatic Therapy and tDCS). Specifically, for gait speed under normal conditions, a significant effect was observed. Gait speed under dual-task conditions also demonstrated a significant intervention effect. Additionally, proprioception of the knee showed a highly significant effect across interventions.

Post-hoc pairwise comparisons revealed further significant differences between groups (Table 4). For gait speed under normal conditions, the Aquatic Therapy + Real tDCS group demonstrated significantly greater gait speed compared to both the Real tDCS group and the Sham tDCS group. Similarly, the Aquatic Therapy + Sham tDCS group showed significantly higher gait speed than both the Real tDCS group and the Sham tDCS group. However, no significant difference was found between the Aquatic Therapy + Real tDCS and Aquatic Therapy + Sham tDCS groups.

Table 4.

Multiple comparisons between groups and the variables of impact of proprioception of knee and gait speed

Variable	Group I	Group J	Mean Difference (I-J)	Std. Error	p	95% Confidence Interval
						Lower Bound	Upper Bound
Gait Speed under normal conditions	Aquatic Therapy + Real tDCS	Aquatic Therapy + Sham tDCS	− 0.293	0.487	0.550	−1.267	0.681
		Real tDCS	−1.332*	0.536	0.016	−2.404	− 0.261
		Sham tDCS	−2.361*	0.504	0.001	−3.369	−1.354
	Aquatic Therapy + Sham tDCS	Real tDCS	−1.039*	0.458	0.027	−1.956	− 0.123
		Sham tDCS	−2.068*	0.446	0.000	−2.960	−1.176
	Real tDCS	Sham tDCS	−1.029*	0.453	0.027	−1.934	− 0.124
Gait Speed in Dual Task	Aquatic Therapy + Real tDCS	Aquatic Therapy + Sham tDCS	0.196	0.608	0.748	−1.019	1.411
		Real tDCS	− 0.590	0.677	0.387	−1.943	0.763
		Sham tDCS	−1.888*	0.627	0.004	−3.142	− 0.635
	Aquatic Therapy + Sham tDCS	Real tDCS	− 0.786	0.571	0.174	−1.927	0.356
		Sham tDCS	−2.084*	0.556	0.001	−3.195	− 0.973
	Real tDCS	Sham tDCS	−1.299*	0.563	0.024	−2.424	− 0.173
proprioception of knee	Aquatic Therapy + Real tDCS	Aquatic Therapy + Sham tDCS	−2.397*	0.118	0.000	−2.632	−2.161
		Real tDCS	−2.667*	0.126	0.000	−2.919	−2.415
		Sham tDCS	−2.780*	0.116	0.000	−3.011	−2.549
	Aquatic Therapy + Sham tDCS	Real tDCS	− 0.270*	0.104	0.012	− 0.479	− 0.062
		Sham tDCS	− 0.383*	0.104	0.000	− 0.591	− 0.176
	Real tDCS	Sham tDCS	− 0.113	0.103	0.276	− 0.319	0.093

*Statistically significant at p ≤ 0.05

For gait speed under dual-task conditions, the Aquatic Therapy + Real tDCS group did not show a significant difference compared to the Aquatic Therapy + Sham tDCS or Real tDCS groups. However, it significantly outperformed the Sham tDCS group. Similarly, the Aquatic Therapy + Sham tDCS group showed significantly greater gait speed compared to the Sham tDCS group. The Real tDCS group also demonstrated a significant improvement over the Sham tDCS group. No other significant differences were observed between the active intervention groups.

In terms of knee proprioception, all intervention groups performed significantly better than the Sham tDCS group. The Aquatic Therapy + Real tDCS group showed superior outcomes compared to both the Real tDCS group and the Aquatic Therapy + Sham tDCS group. Additionally, the Aquatic Therapy + Sham tDCS group demonstrated a small but statistically significant improvement compared to the Real tDCS group. No significant difference was found between the Real tDCS group and the Sham tDCS group.

Discussion

The aim of this randomized controlled trial was to investigate the individual and combined effects of Aquatic Therapy + Real tDCS and Transcranial Direct Current Stimulation (tDCS) on proprioceptive deficits and gait speed impairments in older adults with knee osteoarthritis. The findings demonstrated that the combination of Aquatic Therapy + Real tDCS significantly improved knee proprioception compared to either intervention alone or Sham tDCS, suggesting a synergistic effect. All intervention groups: Aquatic Therapy + Sham tDCS, Real tDCS, and Aquatic Therapy + Real tDCS, exhibited significant improvements in proprioception, whereas the Sham tDCS group showed no meaningful change. Notably, the Aquatic Therapy + Real tDCS group achieved the greatest improvement, outperforming both of its individual components. Regarding gait speed, significant improvements under both normal and dual-task conditions were observed in all intervention groups except the Sham tDCS group. Post-hoc comparisons revealed that, under normal gait conditions, both the Aquatic Therapy + Real tDCS and Aquatic Therapy + Sham tDCS groups achieved superior gait speed compared to the Real tDCS and Sham tDCS groups. However, no significant difference was found between the two Aquatic Therapy groups. Under dual-task conditions, all three active intervention groups significantly outperformed the Sham tDCS group, but no significant differences were observed among the active treatments themselves. These results suggest that while both interventions contribute to gait improvement, Aquatic Therapy may play a more dominant role in enhancing gait performance, particularly under cognitively demanding conditions.

The inclusion of dual-task gait speed aimed to assess gait performance under circumstances that more closely mimic real-world walking, where cognitive and motor demands often coexist [52, 53]. Dual-task assessments are especially useful in uncovering subtle deficits in cognitive-motor integration that may not manifest under single-task walking [54]. Because individuals with knee OA tend to have impairments in both executive function and motor control, testing gait under conditions of cognitive load is more clinically relevant and ecologically valid as an assessment of functional mobility and fall risk [55]. The incorporation of dual-task gait speed thereby enabled us to ascertain whether the interventions not just enhanced physical capacity but also improved the coordination between the motor and the cognitive to provide safe and adaptive walking in everyday life [53].

There is an apparent trend of improvement in knee proprioception in the group which underwent both aquatic therapy and real tDCS, which may imply synergistic effect of the two interventions. This is consistent with existing literature to the effect that both tDCS and aquatic therapy, when individually applied, are effective in improving motor and sensory functions. For instance, tDCS has previously been shown to enhance proprioception and gait performance in different populations through the modulation of the excitability of the brain and sensorimotor integration [29, 51, 56]. In the same vein, aquatic therapy utilizes the concepts of buoyancy, resistance, and hydrostatic pressure to unload joints and provide proprioceptive input to enhance the overall motor control and mobility [57, 58].

The present findings support Coons et al.‘s results which showed that an eight-week underwater treadmill program led to significant improvements in patient-reported outcomes such as stiffness and walking ability among patients with KOA [23]. Assar et al.‘s research shows that women with KOA benefited from aquatic training through enhanced balance, reduced pain, and better joint stability when using TRX or training alone, but joint strength improved more with TRX [59]. Similarly, de Mattos et al. demonstrated that elderly women with KOA experienced better muscular strength, mobility, and quality of life outcomes from high-speed aquatic exercises [58]. Aquatic therapy shows proprioceptive advantages not just for osteoarthritis patients as evidenced by Kum and Shin’s findings of post-stroke patients gaining proprioception after aquatic exercises [60] and Jain and Shinde’s observation of similar gains in people with bilateral KOA [24]. Finally, Azizi et al. reported reductions in pain and improvements in balance and gait parameters following aquatic interventions in elderly patients with KOA [61]. The combined research findings support the idea that aquatic therapy improves proprioception through mechanical and sensory pathways and that tDCS enhances these benefits via neurophysiological adjustments which explain the better results in the group receiving both treatments. These findings confirm that aquatic therapy was primarily responsible for the observed improvements in gait speed under both normal and dual-task conditions. The addition of tDCS did not appear to augment the benefits provided by aquatic therapy alone. This is consistent with the known properties of aquatic therapy, including buoyancy, resistance, and hydrostatic pressure, all of which contribute to reduce joint loading, enhanced muscular activation, improved sensory input, and better neuromuscular coordination [24, 62]. These combined mechanisms likely underlie the improvements in both gait speed and knee proprioception observed in the aquatic therapy groups.

Despite the hypothesis by Teixeira et al. that the combination of tDCS and physical activity has additive effects on the rehabilitation of osteoarthritis [34], the present trial found no significant additive improvements in gait performance when tDCS and aquatic therapy are combined, as compared to aquatic therapy alone. This trend is consistent with results from stroke rehabilitation literature where double-tDCS plus task-specific training did not surpass training alone, though both treatment conditions improved gait performance [63]. This finding suggests that while tDCS shows promise, its additive effects may be highly context-dependent. This context dependency is influenced by several factors identified in previous research, including the individual’s baseline level of motor or cognitive function [64, 65], the stimulation parameters such as current intensity, duration, electrode montage [66, 67], and the cumulative dose or frequency of sessions administered [68]. For instance, higher baseline impairment may result in more pronounced benefits due to greater neural plasticity potential [65], whereas variations in electrode placement and intensity can differentially modulate cortical excitability [66]. These characteristics critically influence treatment response, and further research is needed to systematically investigate these moderators to optimize the efficacy of tDCS interventions.

Although the group receiving both treatments showed the best proprioceptive gains, their improvements in gait speed were not significantly superior to aquatic therapy alone. This suggests a stronger effect of tDCS on sensory processing rather than on motor execution. This difference may stem from the complex nature of the motor function, which encompasses, along with central integration of the senses, the peripheral neuromuscular dynamics and biomechanical restrictions less directly modulated by the stimulation of the cerebral cortex. Chowdhury showed that peripheral nerve stimulation activates subcortical and pre-perceptual pathways, thereby facilitating motor planning and execution more effectively than sensory feedback relying solely on visual or cortical input [69]. Suminski et al. found that both the dorsal and ventral premotor cortices contain neurons that are tuned to limb biomechanics and are cortically sensitive to peripheral dynamics [70]. Similarly, Aizawa et al. found that the primary motor cortex (M1) is responsive to peripheral feedback during the performance of movements [71], highlighting the integrated contribution of sensorimotor feedback to the control of motion. As such, although cortical regions are essential to the initiation and modulation of movements, the actual production of the motor action is rooted in the peripheral system which is less susceptible to modulation from non-invasive cortical stimulation such as tDCS. Anodal tDCS has been shown to enhance cortical excitability and improve sensorimotor integration, particularly through effects on the somatosensory cortex and adjacent proprioceptive areas [72, 73]. However, gait speed, especially in older adults with KOA, is largely dependent on peripheral neuromuscular function, joint mechanics, and biomechanical limitations that are less responsive to cortical modulation [74, 75]. While tDCS has been reported to improve strength and balance under certain conditions [46, 76–78], the distinct biomechanical and therapeutic properties of aquatic therapy, such as buoyancy, resistance, and multisensory engagement likely played a more dominant role in facilitating improvements in motor function. Thus, consistent with prior literature, our findings suggest that tDCS may be more effective in enhancing sensory discrimination and sensorimotor coordination than in directly augmenting gross motor performance [73, 79]. The superior proprioceptive outcomes observed in the aquatic therapy + real tDCS group further support the hypothesis that tDCS exerts neuromodulatory effects on sensory circuits. This is consistent with the work of Rosenkranz et al. [80], who reported enhanced sensory discrimination linked to tDCS-induced cortical plasticity. By modulating cortical excitability and facilitating synaptic plasticity, tDCS may improve sensorimotor integration, a central mechanism underpinning proprioceptive accuracy [81]. Moreover, tDCS may influence descending pain-modulatory pathways and mitigate maladaptive central sensitization, a common feature of chronic pain conditions such as knee osteoarthritis [82]. These neurophysiological adaptations may also involve biochemical mediators, including increased brain-derived neurotrophic factor (BDNF) expression and GABAergic modulation, which together enhance signal integration and motor coordination [83]. The results of this trial are consistent with previous findings by Corrêa et al. [84], Jung et al. [85], and Xiao et al. [32], all of whom reported additive effects of tDCS when combined with physical interventions, particularly in domains related to sensory modulation and sensorimotor control.

Limitations and future direction

This study presents several limitations that should be acknowledged. First, the sample consisted exclusively of female participants, limiting the generalizability of the findings. Future research should include male participants to explore potential gender differences in responsiveness to combined aquatic therapy and tDCS interventions. Second, the study focused solely on short-term outcomes. While improvements in proprioception and gait speed were observed, the durability of these effects remains unknown. Longitudinal studies are needed to assess the long-term sustainability and potential cumulative benefits of the interventions over extended periods. Another methodologic concern is the measurement of dual-task gait performance. Although a 1-back cognitively demanding task was used to induce cognitive-motor interference, no performance on the cognitive task itself (e.g., response time or accuracy) was measured or analyzed. Thus, it is unknown whether gait-speed improvements were the result of optimized gait adjustments at the cost of compensatory reductions in cognitive function. Future studies should include objective metrics of both cognitively and motorically active task performance to further assess dual-task cost and interaction between the two in the population. The current study did not reveal synergistic effects of tDCS and aquatic therapy on gait speed, but it does not exclude the possibility of optimization. The future studies need to examine the differences in tDCS parameters—like intensity, position of the electrodes, and the duration of stimulation—to maximally benefit from the use of tDCS along with physical therapies. It is also possible that neuroimaging methodologies be employed to analyze the effects of tDCS and aquatic therapy on the underlying neural mechanisms, specifically on sensorimotor integration and neuroplasticity. Finally, the combination of tDCS with other physical interventions, such as balance or resistance training, warrants exploration. Such combinations may yield additional insights into multimodal rehabilitation strategies aimed at maximizing functional outcomes in older adults with knee osteoarthritis.

Conclusion

This clinical trial aimed to evaluate the combined effects of aquatic therapy and transcranial direct current stimulation (tDCS) on knee proprioception and gait speed in older adults with knee osteoarthritis (KOA). While both aquatic therapy and tDCS have demonstrated efficacy as standalone interventions, the findings of this study indicate that aquatic therapy alone appears effective in improving gait speed under both normal and dual-task conditions. However, the significant enhancement in knee proprioception observed in the combined intervention group suggests that tDCS may contribute meaningfully to sensory improvements beyond what aquatic therapy alone can achieve.

These results imply that tDCS may facilitate sensory and motor processing through mechanisms related to cortical modulation and sensorimotor integration, particularly in conditions characterized by proprioceptive deficits such as KOA. Although no additive effect of tDCS was observed for gait speed, its potential role in augmenting proprioceptive function highlights the value of further exploring neuromodulation strategies in multimodal rehabilitation protocols.

Authors’ contributions

“Zahra Bagherpoor Kalkhoran: Methodology/Study design, Data curation, ResourcesRazieh Khanmohammadi *: Methodology/Study design, Investigation, Writing– original draft, Review and editing”.

Funding

This research was extracted from a Master's thesis at Urmia University, Iran. No specific funding was received from any external or internal funding agency for this study.

Data availability

“The data generated during this study are summarized in this published article. The full dataset is available from the corresponding author upon reasonable request.”

Declarations

Ethics approval and consent to participate

This study was conducted in accordance with the principles of the Declaration of Helsinki. The study was approved by the Ethics Committee of Urmia University (Approval ID:(IR.URMIA.REC.1402.026)). It was retrospectively registered as a clinical trial on the… Registry of Clinical Trials (IRCT ID: (IRCT20190908044722N8)) on January 8, 2025, due to administrative delays unrelated to the study design or execution. All participants provided written informed consent prior to participation, with the assurance that they could withdraw from the study at any time. This trial adhered to the CONSORT 2010 guidelines to ensure rigorous reporting and transparency.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.