Fatigue-induced changes in static and dynamic balance control in athletes with chronic ankle instability: a systematic review

Abstract

Background

Impaired balance control is considered as a key determinant of chronic ankle instability (CAI). Fatigue can exacerbate sensorimotor deficits in athletes with CAI. However, the effects of fatigue on static and dynamic balance control in athletes with CAI remain unclear. This systematic review aimed to synthesize the existing evidence regarding the effects of fatigue on static and dynamic balance control in athletes with CAI.

Methods

This review included experimental, quasi-experimental, and observational studies. PubMed, Scopus, Web of Science, the Cochrane Library, and Google Scholar were searched from their inception to November 2025. Methodological quality was independently assessed by two reviewers using the Mixed Methods Appraisal Tool (MMAT). Keywords such as “fatigue,” “chronic ankle instability,” and “balance control,” along with their related terms, were used in the search strategy. Findings were synthesized qualitatively.

Results

Nine studies with high methodological quality (mean score: 4.11/5) were included in this review. Dynamic balance was assessed using the Star Excursion Balance Test (SEBT) in three studies, the modified SEBT in one study, the Y-Balance Test in three studies, and the unilateral landing tasks on a force platform in two studies. Static balance was assessed using the Romberg test in one study and the single-leg standing tasks on a force platform in two studies. This review consistently demonstrated the deleterious effects of different fatigue protocols on static and dynamic balance control in athletes with CAI, except for one study that the results showed no significant effect of fatigue on dynamic balance control. Moreover, two studies reported significantly poorer SEBT performance following fatigue in athletes with CAI compared to healthy controls.

Conclusion

The findings suggest that fatigue can adversely affect both static and dynamic balance control in athletes with CAI. However, the extent to which CAI increases susceptibility to fatigue-related balance impairments remains unclear. Future research employing sport-specific fatigue protocols and sensitive balance measures while accounting for relevant participant characteristics may help clarify the effects of fatigue on balance performance in athletes with CAI and inform more targeted rehabilitation and injury-prevention strategies.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s13102-026-02036-6.

Introduction

Lateral ankle sprain is one of the most prevalent sport-related musculoskeletal injuries and is associated with a high rate of recurrence [1]. It is estimated that up to 40% of athletes with acute lateral ankle sprains develop to chronic ankle instability (CAI) [2]. Impaired balance control is considered as a key determinant of CAI [3], which could adversely affect functional performance during sport-specific activities and increase the risk of recurrent ankle sprains [4, 5]. Therefore, examining the static [6] and dynamic balance control [7] has been regarded as a necessity for decision about return to sport following the lateral ankle sprain. These findings highlight the need for a comprehensive rehabilitation approach that not only consider clinical impairments related to ankle joint, but also potential secondary complications, such as balance control to facilitate a holistic return to pre-injury status. In support to this hypothesis, improving balance performance recommended as a priority in the interventions aimed to prevent recurrent sprains in athletes with CAI [8].

In individuals with CAI, impaired static and dynamic balance control is mainly linked to deficiencies in sensory systems and muscular performance [9]. The impaired proprioceptive nerve endings within the sprained joint could cause to the delivery of inaccurate afferent feedbacks to the nervous system, disrupting the body’s balance control [10]. On the other hand, muscular weakness, delayed activation of stabilizing muscles, and decreased coordination between these muscles may limit effective corrective responses and subsequently, disturb balance control [11]. In addition to these peripheral mechanisms, recent evidence have highlighted the potential contribution of central mechanisms for balance deficits in athletes with CAI, including decreased corticospinal excitability of the fibularis longus, tibialis anterior [12], and soleus muscles [13] and structural changes within the central nervous system (CNS) regions involved in balance control [14]. Notably, most sport-related injuries occur during the second half of sports games, especially in the last 15 min when athletes are more likely to experience fatigue [15, 16], with approximately 40% of these injuries involving the ankle joint [17]. Thus, the effects of fatigue on balance control cannot be ignored, especially in those with recurring injuries, such as athletes with CAI. In this context, fatigue may exacerbate balance deficits in athletes with CAI by adversely affecting the sensory and neuromuscular mechanisms involved in balance control. Specifically, fatigue may reduce proprioceptive acuity, diminish muscle force production, and increase delays in neuromuscular responses, specifically in the peroneal muscles [18]. Importantly, these mechanisms are often attributed to peripheral fatigue [19], fatigue may also origin from central mechanisms. Alterations in the nervous system (spinal and supraspinal) under fatigued conditions may induce a failure of motoneuron excitation, resulting in disturbed balance performance [20]. Given the potential effects of peripheral and central fatigue mechanisms on sensorimotor control, their contribution to the persistence of CAI warrants consideration, as athletes with CAI may experience greater impairments in athletic performance and a increased risk of recurrent ankle sprains under fatigued conditions compared to healthy controls [21].

An increasing number of studies have investigated the effects of fatigue on balance control in athletes with CAI. There are numerous variations in the research methods reported to date, including differences in participant’ characteristics, fatigue protocols, and balance assessment tasks. Previous systematic reviews in healthy athletes have suggested that different fatigue protocols do not consistently induce changes in balance performance [22, 23]. Furthermore, it remains unclear whether fatigue affects static and dynamic balance control to a similar extent or whether specific aspects of balance control are more vulnerable to fatigue in athletes with CAI. Despite several systematic review studies investigated balance control and its associated factors in athletes with CAI [24–27], to the best of our knowledge, no systematic review has yet synthesized the current evidence regarding the effects of fatigue on balance control in this population. Given the critical role of balance control in sport-specific performance and the potential adverse effects of fatigue on athletic performance and recurrent injury risk, this systematic review aimed to gather and synthesize the existing evidence regarding the effects of fatigue on static and dynamic balance control in athletes with CAI.

Methods

Protocol and registration

This systematic review was conducted in accordance with the ethical recommendations of Declaration of Helsinki and was approved by the University’s Institutional Review Board (IR.AJUMS.REC.1405.008). Additionally, this study was planned and reported following the recommendations of the PRISMA 2020 statement for systematic reviews [28]. The protocol was registered in advance in the PROSPERO (registration ID: CRD420251237123), providing a publicly accessible record of the intended methods [29].

Eligibility criteria

The PICOS framework was used in this study. This framework comprises five components: Population, Intervention/Exposure, Comparator, Outcomes, and Study design.

Population

Studies were included if they involved athletes with CAI, defined according to the International Ankle Consortium Position Statement as having: (1) a history of at least one significant ankle sprain, with the first sprain occurring at least 12 months before the study; and (2) participants’ reports of giving way in the ankle joint and/or recurrent ankle sprains (two or more sprains to the injured ankle) and/or a feeling of instability confirmed by a Cumberland Ankle Instability Tool (CAIT) score ≤ 24 [30]. Studies were excluded if they (1) examined only isolated acute ankle sprain episodes without chronic or recurrent ankle instability, or (2) did not include athletes as the primary target population.

Intervention / exposure

Local fatigue protocols specifically targeted lower limb muscles or general fatigue protocols were induced to the whole body in a sporting or functional context. Fatigue was mainly identified as the inability to continue performing the fatiguing task and was confirmed by Borg’s rating of perceived exertion, heart rate or objective performance-based criteria [31].

Comparator

Pre-fatigue measurements, non-fatigued control conditions.

Outcome measures

Assessment of balance control—either static or dynamic—was required for inclusion. Static balance control was assessed using laboratory-based measures, such as force platform during single-leg standing (variables: center-of-pressure variables, time-to-boundary (TTB)), or standardized clinical tests (e.g., Romberg test, single-leg stance). Dynamic balance control was assessed using laboratory-based measures, such as force platform during unilateral landing tasks (variables: ground reaction forces, time to stabilization (TTS)), or standardized clinical tests (e.g., Y-Balance Test (YBT), Star Excursion Balance Test (SEBT), modified SEBT). Included studies assessed balance control before and after the fatigue protocol, allowing evaluation of fatigue-related changes in balance control.

Study design

Experimental, quasi-experimental, case–control, cross-sectional studies, and cohort studies were eligible for inclusion. Studies were excluded if they were study protocols, review articles, opinion papers, or editorials.

Information sources and search strategy

A comprehensive electronic search was carried out in four major databases: PubMed, Scopus, Web of Science and the Cochrane Library. All databases were searched from their inception up to November 20, 2025. In addition, Google Scholar was used as an additional source to identify further relevant studies through forward citation tracking and screening of related records. The search strategy was as follows: (“ankle injuries” OR “chronic ankle instability” OR “chronic lateral ankle instability” OR “ankle instability” OR “ankle joint instability” OR “lateral ankle instability” OR “functional ankle instability” OR “mechanical ankle instability” OR “recurrent ankle sprain”) AND (“fatigue” OR “muscle fatigue” OR “motor fatigue” OR “physical fatigue” OR “exercise-induced fatigue” OR “exhaustion” OR “peripheral fatigue” OR “local fatigue” OR “functional fatigue” OR “whole-body fatigue”) AND (“postural balance” OR “balance” OR “equilibrium” OR “postural control” OR “postural stability” OR “postural sway” OR “posturography” OR “stabilometry” OR “force platform analysis”). Specific search strategies for each database are reported in Appendix A.

Selection process

Two reviewers (M.M.A. and R.M.) independently screened the titles and abstracts of all records against the predefined eligibility criteria. Inter-rater agreement was evaluated using the percentage of agreement and Cohen’s kappa coefficient. The Rayyan web-based tool was used to screen titles and abstracts and to classify studies during the selection process [32]. Full texts of potentially relevant articles were then retrieved and assessed independently by the same two reviewers. Any disagreements at either stage were first resolved through discussion; if consensus could not be reached, a third reviewer (N.N.A.) acted as an arbitrator. During the study selection process, only studies published in English were considered eligible and included in the review.

Data collection process and data items

Data extraction was conducted independently by two reviewers (M.M.A. and R.M.) using a standardized spreadsheet. Discrepancies were resolved through discussion and, when necessary, consultation with a third reviewer (N.N.A.). For each included study, the following data were extracted: first author and year of publication, study design, participant characteristics, fatigue protocol, testing procedure, outcome measures, and main findings.

Quality assessment

Methodological quality of the included studies was appraised using the Mixed Methods Appraisal Tool (MMAT) [33]. This instrument allows the use of a single framework to evaluate different quantitative study designs, which was appropriate for the range of designs included in this review [33, 34]. Two initial screening questions were applied to confirm that studies were empirical and had clearly stated objectives and outcomes. For each eligible study, five design-specific criteria were rated using a binary scoring system (1 = criterion met; 0 = criterion not met or could not be determined). The overall quality score for each study therefore, ranged from 0 to 5. Following the quality classification approach adopted in prior literature [35], scores of 4–5 were interpreted as high methodological quality, a score of 3 as moderate quality, and scores of 0–2 as low quality. Quality assessment was conducted independently by two reviewers (M.M.A. and R.M.), with any disagreements resolved through consultation with a third reviewer (N.N.A.). Inter-rater agreement was evaluated using the percentage of agreement and Cohen’s kappa coefficient. All studies were retained in the review regardless of their MMAT score, and quality ratings were considered when interpreting the results.

Data synthesis

In this review study, a meta-analysis was not conducted due to the heterogeneity among the included studies concerning participant’ characteristics, fatigue-induction protocols and criteria for its confirmation, and balance assessment tasks. Instead, a structured narrative synthesis was undertaken to summarize and integrate the findings.

Results

Study selection

In total, 227 records were identified through all search sources. After the removal of 77 duplicates, 150 unique records remained for title and abstract screening. Inter-rater reliability during the study selection process yielded an overall agreement rate of 95.33% and a Cohen’s kappa coefficient of 0.83. Following this stage, 23 articles were retrieved for full-text assessment. Fourteen full-text articles were excluded for the following reasons: full text not available in English (n = 2), absence of an experimentally induced fatigue protocol (n = 1), lack of balance assessment (n = 1), and populations not restricted to athletes with CAI or without separate analyses for this subgroup (n = 10). Consequently, 9 studies met the predefined inclusion criteria and were included in the final qualitative synthesis (Fig. 1).

Fig. 1.

PRISMA diagram showing the flow of studies through phases of the review

Study characteristics

A total of 9 studies, encompassing 296 participants, were included in the final review. Of the 9 included studies, five employed a quasi-experimental design [21, 36–39], two used a pre–post experimental design [40, 41], one adopted a cross-sectional design [42], and one was conducted as a cohort study [43]. Study characteristics and statistical findings are presented in Tables 1 and 2, respectively.

Table 1.

Characteristics of the included studies examining the effect of fatigue on static and dynamic balance control in athletes with CAI

First author (year)	Country	Study design	Study population	Fatigue protocol	Testing procedure	Outcome measures	Main findings
Kim (2024) [42]	Korea	Cross-sectional and repeated measures	28 young physically active individuals with CAI (mean age: 24.07 years)	Agility exercises include: 5 m forward dashes, side dashes, backward dashes, L-shaped running, repetitive side jumps, forward countermovement jumps, and maximum jumps while bringing both knees up to the chest	10 s single-leg standing on the force platform with EO and EC	TTB minima means and SD in both ML and AP directions, as well as %modulation of these variables	Fatigue significantly impaired balance control with EO but not EC and decreased visual reliance in athletes with CAI
Dury (2024) [36]	France	Quasi-Experimental	23 physically active males were divided into two groups: 13 healthy controls (mean age: 22.9 years) and 10 with CAI (mean age: 23.5 years)	A circuit of four agility exercises was performed: 10 lateral-slide around 2 cones (an 8-configuration with cones spaced at 2 m); 20 lateral-side crossing over a step (20 cm height); 30 plyometric single-leg forward/backward jumps; 60 vertical bipedal bounded jumps. This sequence was repeated 3 times and then a plantar flexion strength measurement was performed. The fatigue session continued until a decrease in plantar flexion strength of 20% was observed.	Unilateral landing task on the force platform during single and dual-task conditions (two-dimensional visual object tracking)	GRF and TTS in vertical, AP and ML directions, as well as, three-dimensional movements of the hip, knee and ankle joints	No significant interaction were found between factors (group, fatigue, cognitive). Neuromuscular fatigue led to impaired landing biomechanics by reinforcing an ankle protection strategy and inducing proximal joints to compensate. Furthermore, in response to fatigue, dynamic balance control during landing task was significantly decreased.
Muhammad (2023) [40]	Malaysia	Pre- and post-experimental study	36 elite youth netball players with CAI (mean age: 1.76 years)	SSC protocol consisting of 10 sets of 10 repetitions of consecutive maximal bilateral CMJs (30-s rest between sets) on the jump mat.	YBT	Maximum reach distance	No significant effect of fatigue on dynamic balance control
Azhar (2023) [37]	Malaysia	Quasi-Experimental	28 elite youth soccer players with CAI were divided into two groups: 14 with ankle taping (age: 18.93±0.829) and 14 control without taping (age: 19±1.301).	YoSFS5 fatigue protocol:5 minutes of circuit involving several soccer-specific movements (kicking,jumping, and heading, sprints with a ball)	YBT	Maximum reach distance and YBT composite score	Fatigue decreases maximum reach distances in PL and PM directions and increases the YBT composite score.
Bagherian (2018) [41]	Iran	Pre- and post-experimental study	40 male collegiate athletes with CAI (age: 21.02 ± 1.7 years)	Treadmill walking at a speed of 6 km/h. Treadmill speed then increased 1 km every 2 min until the participant’s perceived exertion, using the Borg scale, reached 17 (very hard) or the patient’s maximum heart rate reached 90% of their age adjusted max heart rate	Modified SEBT	Maximum reach distance	Fatigue significantly affected dynamic balance control in athletes with CAI
Yalfani (2017) [21]	Iran	Quasi-Experimental	24 physically active females were divided into two groups: 12 with CAI (age: 22±3.8 years), 12 healthy controls (age: 21.3±3.7 years)	10-min warm-up followed by hip abduction, squats, lateral band walks, and multiplanar lunges/hops in the sagittal, frontal, and transverse planes (gluteal muscles fatigue protocol).	Romberg test (EO, EC) for assessment of static balance, YBT for assessment of dynamic balance	Romberg score, Maximum reach distance in YBT	Fatigue significantly decreased balance scores for both groups with no significant group differences
Steib  (2013) [43]	Germany	Cohort study	57 athletes; age range 18 to 35 years; divided into three groups: 19 FAI (age: 24.95±3.17), 19 coper (age: 24.53±2.76), 19 healthy controls (agr: 23.32±3.79).	Running exercise on a motorized treadmill until subjective exhaustion.  a RPE of 17 (“very hard”) and higher defined as a fatigued state	20s single leg stance on the force platform for assessment of static balance control, SEBT in Anterior, posterior, medial and lateral directions and unilateral jump landing task for assessment of dynamic balance control	COP velocity in AP and ML directions for assessment of static balance control, Maximum reach distance, Time to stabilization in AP and ML directions for assessment of dynamic balance control	Fatigue caused similar effects on static balance control in all three groups. For dynamic balance control, fatigue had also a diminishing effect in all three group. However, copers responded differentlyto fatigue in anterior and medial reach directions compared to theFAI and control groups.
Gribble (2007) [38]	USA	Quasi-Experimental	30 physically active individuals were divided into two groups: 14 CAI (7males, 7 females; age: 21.9 ± 2.9 years), 16 healthy controls (8 males, 8    females; age: 22.5 ± 2.4 years)	Two conditions: 1) Isokinetic ankle fatigue (plantarflexion/dorsiflexion), 2) Lunging task (dynamic fatigue protocol for the lower extremity).	SEBT in anterior, medial, and posterior directions.	Sagittal-plane kinematics of the stance leg; maximum reach distance	CAI and fatigue influenced dynamic balance control, altering knee and hip flexion angles, under lunge fatigue but not isolated ankle muscles fatigue.
Gribble (2004) [39]	USA	Quasi-Experimental	30 physically active individuals were divided into two groups: 14 CAI (7males, 7 females; age: 21.9 ± 2.9 years), 16 healthy controls (8 males, 8    females; age: 22.5 ± 2.4 years)	Five condition: control condition (quiet sitting), fatigue inducing by a Biodex System III isokinetic dynamometer for hip, knee, and ankle joints, fatigue inducing by performing maximum lunging task	SEBT in anterior, medial, and posterior directions.	Sagittal-plane kinematics of the stance leg; maximum reach distance	Chronic ankle instability and fatigue disrupted dynamic postural control, most notably by altering control of sagittal-plane joint angles proximal to the ankle.

CAI Chronic Ankle Instability, SD Standard deviation, SEBT Star Excursion Balance Test, COP Center of Pressure, YoSFS5 Youth Soccer Fatigue Simulation, HR Heart Rate, EO Eyes open, EC Eyes closed, YBT Y Balance Test, TTB Time to Boundary, ML Mediolateral, AP Anteroposterior, GRF Ground Reaction Force, PL Posterior-lateral, PM Posterior-medial, SSC Stretch-Shortening Cycle, CMJ Countermovement jumps, FAST-FP Functional Agility Short-Term Fatigue Protocol, Normal–Open (N–O) Standing on a firm surface with eyes open, Normal–Closed (N–C) Standing on a firm surface with eyes closed, Cushion–Open (C–O) Standing on a foam surface with eyes open, Cushion–Closed (C–C) Standing on a foam surface with eyes closed, FAI Functional ankle instability

** C90 area: The ellipse area that encompasses 90% of the center of pressure data. It is used as an indicator of static stability in tests with eyes open and eyes closed. An increase in C90 area indicates a decrease in balance control and stability

Table 2.

Statistical findings for fatigue-induced changes in static and dynamic balance control across the included studies

First author (year)	Assessment Tool / Task	Variable	Effect Size	Statistics
Kim (2024) [42]	Single-leg standing on the force platform	TTB ML Mean of minima	0.6	p = 0.009 (Fatigue Effect), F = 7.82
		TTB AP Mean of minima	0.45	p < 0.001 (Fatigue Effect), F = 426.79
		TTB ML SD of minima (s)	0.57	p = 0.091 (Fatigue Effect), F = 63.98
		TTB AP SD of minima (s)	0.54	p < 0.001 (Fatigue Effect), F = 46.88
Dury (2024) [36]	Unilateral landing task on the force platform	Vertical ground reaction force (% BW)	0.43	p < 0.01
		Medial ground reaction force (% BW)	0.31	p < 0.05
		Lateral ground reaction force (% BW)	0. 31	p < 0.05
		Antero-posterior time to stabilization (% BW)	0.53	p < 0.01
Muhammad (2023) [40]	YBT	Anterior Reach	0.13 *	Z = 0.827, p = 0.408
		Posteromedial Reach	-0.06 *	t (35) = − 0.278, p = 0.783
		Posterolateral Reach	-0.004 *	t (35) = 0.019, p = 0.985
Azhar (2023) [37]	YBT	Anterior Reach	-1.4 *	p < 0.001
		Posteromedial Reach	-0.80 *	p < 0.001
		Posterolateral Reach	-0.80 *	p < 0.001
Bagherian (2018) [41]	Modified SEBT	Anterior (%)	0.88	p < 0.001
		Posteromedial (%)	1.76	p < 0.001
		Posterolateral (%)	0.83	p < 0.001
Yalfani (2017) [21]	Romberg test	Romberg test open eyes	0.51	F(1,22) = 38.3, p = 0.0005
		Romberg test close eyes	0.70	F(1,22) = 51.5, p = 0.0005
	YBT	Y test (composite / total score)	0.48	F(1,22) = 20.4, p = 0.0005
Steib (2013) [43]	Single leg stance on the force platform	AP COP velocity, cm/s	FAI group: 1.14 *	p ≤ 0.001
			Healthy group: 1.06 *	p ≤ 0.001
		ML COP velocity, cm/s	FAI group: 0.95 *	p ≤ 0.001
			Healthy group: 1.28 *	p ≤ 0.001
	SEBT	Anterior	FAI group: -0.03 *	p = 0.787
			Healthy group: -0.16 *	p = 0.155
		Medial	FAI group: −0.42	p = 0.002
			Healthy group: −0.29	p = 0.031
		Lateral	FAI group: -0.08 *	p = 0.183
			Healthy group: -0.21 *	p = 0.055
		Posterior	FAI group: -0.10 *	p = 0.360
			Healthy group: -0.13 *	p = 0.485
	Unilateral jump landing on force platform	AP TTS, s	FAI group: 0.37 *	p = 0.011
			Healthy group: 0.33 *	p ≤ 0.001
		ML TTS, s	FAI group: 0.79 *	p ≤ 0.001
			Healthy group: 0.53 *	p ≤ 0.001
Gribble (2007) [38]	SEBT	Normalized reach distance (MAXD, % leg length)	NC	p < 0.05
Gribble (2004) [39]	SEBT	Normalized reach distance (MAXD, % leg length)	NC	p < 0.001

Effect sizes marked with an asterisk (*) were not reported in the original studies and were calculated by the authors. For parametric data, effect size was calculated as (Meanpost−Meanpre)/SD_pre [44]. For non-parametric data, effect size was calculated as Z-value divided by the square root of N [45]

Abbreviations: AP anteroposterior, BW body weight, CoP center of pressure, F F-statistic, FAI functional ankle instability, MAXD maximum reach distance, ML mediolateral, p probability value, SD standard deviation, SEBT Star Excursion Balance Test, t t-statistic, TTB time to boundary, TTS time to stabilization, YBT Y-Balance Test, Z Z-statistic, NC Not calculable.

Quality assessment

Using the MMAT scores, 1 study achieved a score of 5 [38],and 8 studies scored 4 [21, 36, 37, 39–43]. Accordingly, all studies were classified as having high methodological quality (scores 4–5) (Table 3 and Fig. 2). Inter-rater reliability for the methodological quality assessment yielded an overall agreement rate of 95% and a Cohen’s kappa coefficient of 0.83. Following additional discussions, they reached complete agreement.

Table 3.

Methodological quality assessment of included studies using the mixed methods appraisal tool (score and quality category)

First Author (Year)	Are the participants representative of the target population?	Are measurements appropriate regarding both the outcome and intervention (or exposure)?	Are there complete outcome data?	Are the confounders accounted for in the design and analysis?	During the study period, is the intervention administered (or exposure occurred) as intended?	MMAT Score (0–5) & Quality Category
Kim (2024) [42]	Yes	Yes	Yes	No	Yes	4 (High)
Dury (2024) [36]	Yes	Yes	Yes	No	Yes	4 (High)
Muhammad (2023) [40]	Yes	Yes	Yes	No	Yes	4 (High)
Azhar (2023) [37]	Yes	Yes	Yes	No	Yes	4 (High)
Bagherian (2018) [41]	Yes	Yes	Yes	No	Yes	4 (High)
Yalfani (2017) [21]	Yes	Yes	Yes	No	Yes	4 (High)
Steib (2013) [43]	Yes	Yes	Yes	No	Yes	4 (High)
Gribble (2007) [38]	Yes	Yes	Yes	Yes	Yes	5 (High)
Gribble (2004) [39]	Yes	Yes	Yes	No	Yes	4 (High)

MMAT Mixed methods appraisal tool

Fig. 2.

Methodological quality assessment of the included studies using the Mixed Methods Appraisal Tool (MMAT). Each study was evaluated across five criteria: (1) representativeness of participants to the target population (2), appropriateness of measurements for both the outcome and intervention (or exposure) (3), completeness of outcome data (4), consideration of confounders in the study design and analysis, and (5) whether the intervention was administered (or exposure occurred) as intended during the study period. Responses to each criterion are indicated as Yes (green), No (red), or Can’t tell (yellow). The overall quality score ranges from 0 to 5 and is categorized as low (0–2), moderate (3), or high quality (4–5)

Fatigue protocols

In eight studies, general fatigue protocols were employed, while local fatigue protocols were utilized in three studies. The local protocols targeted lower-extremity musculature, including isokinetic fatigue of the ankle, knee, and hip [38, 39] and gluteal muscle fatigue [21]. The general fatigue protocols consisted of: (1) functional tasks, such as agility-based drills [36, 42] (n = 2), and lunges [38, 39] (n = 2); (2) plyometric tasks, involving stretch–shortening cycle jump [40]; (3) endurance protocols, including treadmill walking [41] and treadmill running [43]; and (4) sport-specific simulation, such as soccer-specific fatigue circuits [37] (Fig. 3).

Fig. 3.

Fatigue protocol differences

Effect of induced fatigue on static balance control in athletes with CAI

All studies demonstrated the significant effects of fatigue on static balance control in both athletes with CAI and healthy controls. Kim et al. showed that functional fatigue protocol led to a significant decrement in the average values of TTB mediolateral (ML) mean of minima (p = 0.009), TTB anteroposterior (AP) mean of minima (p < 0.001), and TTB AP standard deviation (SD) of minima (p < 0.001) variables in athletes with CAI under single-leg standing with eyes open condition. Furthermore, for the assessment of visual reliance, %modulation scores for each variable were calculated using the following formula: [( eyes open-eyes closed)/eyes open] × 100. Lower values of this variable suggest decreased reliance on visual information and a greater emphasis on somatosensory information for balance control. The results indicated that fatigue reduced %modulation in AP TTBmean (p < 0.001), ML TTBsd (p = 0.05), and AP TTBsd (p < 0.001) in athletes with CAI [42]. Yalfani et al. also observed that Romberg test scores were significantly lower after the induction of gluteal muscle fatigue in both athletes with CAI and healthy controls under open (p = 0.0005) and eye closed conditions (p = 0.0005), with no significant group differences under these conditions (p = 0.50, p = 0.70, respectively) [21]. Steib et al. reported that the three groups of copers, FAI, and healthy control athletes showed greater values of the centre-of-pressure (COP) sway velocity in AP (p ≤ 0.001) and ML directions (p ≤ 0.001) under single-leg standing following fatiguing treadmill running [43].

Effect of induced fatigue on dynamic balance control in athletes with CAI

Among the four studies that used YBT for assessing dynamic balance control, two studies showed that fatigue significantly affected dynamic balance control in both groups of athletes with CAI and healthy controls. Yalfani et al. observed that the gluteal muscles fatigue protocol significantly decreased maximum reach distance from pre-fatigue to post-fatigue in both groups of athletes with CAI and healthy controls (p = 0.0005) [21]. In addition, in a study conducted by Azhar et al., elite youth soccer players with CAI who were allocated into two groups with and without ankle taping showed significantly decreased maximum reach distance in anterior (ANT), posteromedial (PM) and posterolateral (PL) directions and increased the YBT composite score following youth soccer fatigue simulation (YoSFS⁵) (p < 0.01) [37]. Muhammad et al., however indicated no significant effect of fatigue induced by the stretch-shortening cycle on the performance of YBT (p > 0.05) [40].

In all studies that used SEBT/its modified version for assessing dynamic balance control, the results demonstrated the significant effects of fatigue on dynamic balance control in athletes with CAI. Bagherian et al. showed impaired SEBT scores in ANT, PM, and PL directions following fatigue treadmill walking in athletes with CAI (p < 0.001) [41]. Also, Steib et al. found that fatiguing treadmill running caused significantly decreased medial reach distances in the functional ankle instability (FAI) (p = 0.002) and control groups (p = 0.03) and significant decrement in ANT reach distance in copers (p ≤ 0.001). In addition, a significant group-by-fatigue interaction was observed for anterior reach distances. Copers demonstrated larger fatigue-induced reductions compared to FAI (p = 0.007) and control groups (p = 0.05) [43]. Moreover, in a study by Gribble et al., the effects of different fatigue conditions (ankle, knee, hip, lungue fatigue and control conditions with no fatiguing task) on dynamic balance performance (SEBT in ANT, medial, and posterior directions) were compared between athletes with CAI and healthy controls. The results showed that four fatigue conditions had a decreasing effect on maximum normalized reaching distance in the posterior direction in the lower limbs of both groups (p < 0.001), whereas the control day had no significant prefatigue-postfatigue effect. However, larger post-fatigue reduction was observed in the involved limb of CAI group compared to the uninvolved limb of the CAI and healthy control groups (p < 0.001). Additionally, in the posterior direction, hip flexion on the involved limb of the CAI group was significantly lower compared to the the uninvolved limb of the CAI group and the matched limb of the healthy control group (p = 0.02) [39]. In another study, Gribble et al., also investigated the combined effects of CAI and lower extremity fatigue (lunging fatigue protocol and an open chain ankle isokinetic fatigue protocol) on the SEBT performance. The most interesting predictive stepwise model involved the ANT and medial reaching directions. So, when reaching anteriorly after the lunge fatigue condition, CAI and the changes in knee and hip flexion predicted approximately 49% of the variance in % maximum normalized reaching distance (R² = 0.48; p = 0.001). For reaching medially under lunge fatigue, CAI predicted approximately 20% of the variance in %maximum normalized reaching distance (R² = 0.19; p = 0.01). However, isolated ankle fatigue did not cause significantly different responses between groups (p = 0.57, p = 0.08, for ANT and medial reaching directions, respectively) [38].

Additionally, two studies assessed dynamic balance control during the unilateral jump landing task. Steib et al. found that three groups of copers, FAI, and healthy control athletes showed larger fatigue induced increments for TTS in AP (p = 0.04, p = 0.01, and p ≤ 0.001, respectively) and ML directions (p = 0.006, p ≤ 0.001, and p ≤ 0.001, respectively) [43] during unilateral jump landing task. In another study, Dury et al. compared the effects of neuromuscular fatigue and cognitive constraints on lower extremity landing biomechanics between physically active males with CAI and healthy controls. The results demonstrated no significant interaction between groups, fatigue, and cognitive factors for any variable assessed in this study (p > 0.05). Fatigue resulted in decreased ankle plantar flexion (p < 0.001) and inversion (p = 0.04) and increased knee flexion (p = 0.001) in both groups. Moreover, fatigue decreased vertical ground reaction force (p < 0.01) and increased ML GRF (p < 0.05) and AP TTS (p < 0.01) in both groups [36].

Discussion

This systematic review synthesized the current evidence regarding the effects of fatigue on static and dynamic balance control in athletes with CAI. This review indicated the deleterious effects of different fatigue protocols on static and dynamic balance control in athletes with CAI. However, the available evidence did not consistently demonstrate greater fatigue-related impairments in balance performance in athletes with CAI compared to healthy controls.

Regarding to the static balance control, the results of studies showed impaired balance performance following fatigue in athletes with CAI [21, 42, 43]. However, in two studies comparing the differences in response to fatigue between athletes with CAI and healthy controls [21, 43], the results showed no significant group differences. For dynamic balance control, of the 8 articles reviewed in this study, the results of 7 articles indicated the significant influences of various induced fatigue protocols on SEBT/YBT performance or stability during unilateral jump landing task in athletes with CAI [21, 36–39, 41, 43]. Moreover, of the 5 studies comparing fatigue-related alterations in balance performance between athletes with CAI and healthy controls [21, 36, 38, 39, 43], only in 2 studies conducted by Gribble et al., the results showed significant group differences [38, 39]. The relatively consistent observation that fatigue affects static and dynamic balance control in both athletes with CAI and healthy controls suggests that fatigue itself represents a substantial challenge to balance control, whereas the evidence that CAI confers an additional or disproportionate susceptibility to fatigue-induced impairments remains limited.

Given the multifactorial nature of both CAI [46] and fatigue [47], fatigue-related balance impairments are likely to arise from the interplay of multiple contributing factors. In athletes with CAI, impaired sensorimotor integration within complex neural networks has been well established as underlying mechanism for balance control impairments [48]. Interestingly, the researchers suggested that the combined effects of peripheral and central fatigue can also lead to changes in the sensorimotor integration of information involved in balance control [49]. Therefore, fatigue may exacerbate motor control deficits, particularly balance control in athletes with CAI. At the peripheral level, Fatigue-induced changes in somatosensory inputs can potentially lead to deficits in neuromuscular control, resulting in impaired balance performance [21, 43]. While, at the central level, fatigue may affect sensory weighting strategies adopted to maintain balance control [50]. Athletes with CAI are believed to rely more heavily on visual inputs for balance control due to impaired somatosensory function [51]. Surprisingly, Kim et al. indicated a significant decrease in reliance on visual inputs during single leg stance following the induction of fatigue in athletes with CAI [42]. One plausible explanation for this result may be that fatigue-induced alterations in sensorimotor processing may impair the CNS’s capacity to effectively incorporate reliable visual feedback for balance control, thereby reducing the contribution of visual inputs despite its availability. Moreover, Dury et al. reported that fatigue-induced impairments in dynamic stability during landing in athletes with CAI were accompanied by kinematic changes, not only at the injured ankle but also at joints proximal to the ankle [36]. Similarly, Gribble et al. [38, 39] demonstrated that fatigue-related reductions in the SEBT reach performance in the involved limb of athletes with CAI were associated with kinematic alterations at proximal joints of the lower limb. Collectively, these findings suggest that fatigue not only can exacerbates pre-existing ankle deficits in athletes with CAI, but also may alter the movement strategies used across the lower limb kinetic chain to maintain dynamic balance control, potentially increasing the susceptibility of athletes with CAI to further injury. Whether these changes result from central or peripheral fatigue, or more likely, a combination of both, remains unclear. These hypotheses should be investigated in future studies.

A major challenge in interpreting the evidence is the substantial methodological heterogeneity across studies. Differences in participant characteristics, fatigue-induction protocols and confirmation criteria, and balance assessment tasks limit cross-study comparability and the generalizability of the findings. The potential sources of heterogeneity are discussed in detail below.

Heterogeneity in participant’ characteristics

Across the included studies, participants represented a broad range of athletic levels, from recreational to elite athletes, representing an important source of heterogeneity. Elite athletes generally exhibit greater resistance to peripheral and central fatigue due to superior neuromuscular, metabolic, and oxidative adaptations, as well as greater recovery capacity, whereas recreational athletes may be more susceptible to exercise-induced fatigue and slower functional recovery due to lower levels of training adaptation and physiological conditioning [52]. Therefore, athletic level may influence both the development of fatigue and the ability to maintain balance control under fatigue condition. Indeed, findings derived from recreationally active athletes may not be generalizable to elite populations, and vice versa [22]. In this review, fatigue-related impairments in balance control were consistently observed in recreational athletes [21, 36, 38, 39, 42], whereas findings among elite athletes were less consistent [37, 40]. One of the two studies involving elite athletes with CAI reported no significant effect of stretch–shortening cycle jump-induced fatigue on dynamic balance control, potentially reflecting their greater capacity to maintain postural performance under high neuromuscular demands [40]. Future research should determine whether athletic level moderates the effects of fatigue on balance control in individuals with CAI.

Sex may represent an important source of heterogeneity in fatigue-related balance responses, given established sex differences in neuromuscular function [53, 54], balance control [54, 55], and fatigue resistance [56]. Supporting this notion, Hill et al. demonstrated exercise-induced impairments in quiet standing balance among recreationally active men, but not women [57]. However, only three studies in this review accounted for sex [21, 36, 41], highlighting the need for future research to determine whether sex moderates fatigue-related balance impairments in athletes with CAI.

Heterogeneity in fatigue-induced protocols and criteria for its verification

Across the included studies, fatigue was induced using diverse protocols, including local muscle fatigue [21, 38, 39], whole-body endurance activities such as treadmill walking/running [41, 43], and, most commonly, functional fatigue protocols designed to replicate sport-specific activities [36–40, 42]. The latter typically incorporated movements associated with a high incidence of ankle injuries, such as jumping and rapid changes of direction. Notably, Azhar et al. [37] used a sport-specific fatigue protocol simulating youth soccer activity. The diversity of fatigue protocols may potentially affect the fatigue-related changes in balance control, as different protocols impose distinct physiological demands on neuromuscular and sensory systems [58]. Consequently, fatigue-related balance impairments may involve distinct mechanisms, with local muscle fatigue primarily affecting afferent feedback, neuromuscular control, and contractile function, whereas general fatigue may impair balance control through alterations in central integration/processing, physiological homeostasis, and the sensitivity of sensory receptors [59]. Therefore, local fatigue protocols may not fully replicate the fatigue demands experienced during sports participation [60]. Sport activities typically involve repetitive whole-body, multi-joint movements, resulting in general fatigue that can be caused by the combination of peripheral and central mechanisms [61]. Thus, local muscle fatigue protocols may underestimate fatigue-induced alterations in balance control. This hypothesis was supported by the results of a study conducted by Gribble et al. that showed close-chain fatigue protocol (lunge fatigue) rather than isolated ankle muscles fatigue could better clarify dynamic balance control dysfunctions arising from alterations of neuromuscular control in proximal joints of the injured limb in athletes with CAI [38]. Given that balance control strategies vary according to sport-specific demands [22], fatigue protocols incorporating sport-specific characteristics may better reflect real-world athletic conditions and provide greater ecological validity. Further research using validated sport-specific fatigue protocols is warranted.

Moreover, criteria used to confirm fatigue varied from objective physiological thresholds (e.g., heart rate exceeding 90% of maximum) [41, 42], objective performance-based criteria (e.g., a decrease in force production to < 50% of the peak force [38, 39], a 20% reduction in plantar flexion strength [36], a maximum vertical jump height of less than 80% [42], inability to maintain proper lunge form [38, 39], inability to touch the bar [21] to subjective rating scale (e.g., RPE values of 17–20 on the Borg scale) [21, 41–43]. These different fatigue criteria may not reflect equivalent levels of physiological fatigue, which could influence the magnitude of balance impairments and limit comparisons of dose–response relationships across studies.

Heterogeneity in balance assessment tasks

The included studies assessed static balance using single-leg stance on the force platform and the Romberg test, and dynamic balance using the SEBT, YBT, and unilateral jump-landing tasks on force platform. These assessments capture different aspects of balance control and may therefore differ in their sensitivity to fatigue-related impairments in athletes with CAI.

Regarding the sensitivity of balance metrics to fatigue-related impairments, Kim et al. reported a significant deterioration in single-leg standing balance control following a functional fatigue protocol in athletes with CAI, as evidenced by reduced TTB values under the eyes-open condition, whereas no significant pre- to post-fatigue changes were observed with the eyes closed [42]. Conversely, Yalfani et al. reported significant reductions in Romberg test scores following localized gluteal muscle fatigue under both eyes-open and eyes-closed conditions in athletes with CAI [21]. These divergent findings may be attributable to differences not only in the nature and physiological demands of the fatigue protocols but also in the characteristics of the balance assessment tasks used. Collectively, the findings suggest that the ability to detect fatigue-related alterations in balance control may depend on the interaction between the nature of the fatiguing stimulus and the sensitivity of the assessment tool. In particular, force-platform-derived measures may be more sensitive to subtle fatigue-related changes in balance control than clinical balance measures such as the Romberg test.

Moreover, significant group differences were observed in only 2 studies assessing the effects of fatigue on dynamic balance [38, 39], whereas no significant group differences were reported in studies assessing static balance control. On the one hand, this finding may partly support this notion that dynamic balance tasks are more demanding and sport-specific compared to static balance tasks [62]. Thus, they may be more sensitive to identify balance impairments following fatigue in athletes with CAI. On the other hand, in the 2 studies reporting significant group differences, dynamic balance control was assessed using the SEBT. This finding may demonstrated the ability of the SEBT to identify fatigue-related differences in dynamic balance control between athletes with CAI and healthy controls. However, the limited number of studies and methodological heterogeneity preclude definitive conclusions regarding the relative sensitivity or superiority of balance measures for detecting fatigue-related impairments. Further work is warranted to determine the relative sensitivity of different balance measures to fatigue-related impairments.

Limitations and directions for future research

This study has some limitations. Firstly, the number of articles included in this systematic review was low (n = 9), particularly for static balance control (n = 3). Therefore, our data should be interpreted with caution. Secondly, most studies have investigated CAI as a homogeneous condition, leaving it unclear whether fatigue-induced balance impairments differ across clinical subgroups. Future studies should examine subgroup-specific responses to fatigue to better understand mechanisms underlying balance control deficits in athletes with CAI and inform individualized assessment and rehabilitation strategies. Tertiary, the studies concentrated only on the acute effects of fatigue on balance control in athletes with CAI and did not assess the long-term effects. Further research is warranted to determine how long the negative effects of fatigue persist. Furthermore, the results of this review highlight the necessity for performing further research on the mechanisms by which fatigue might contribute to balance dysfunctions.

Clinical implications

Although methodological heterogeneity limits the strength of clinical recommendations, the findings support the potential integration of fatigue-resistance training and fatigue-based balance exercises into injury prevention and rehabilitation programs for athletes with CAI. Additionally, appropriate recovery strategies may help mitigate fatigue-related balance impairments [63]. Cold-water immersion, as a commonly used cryotherapy modality has been reported to facilitate metabolic and perceptual recovery, and attenuate post-exercise performance decrements [64, 65]. however, its effects on fatigue recovery in athletes with CAI require further investigation.

Conclusion

The findings suggest that fatigue can adversely affect both static and dynamic balance control in athletes with CAI. However, the extent to which CAI increases susceptibility to fatigue-related balance impairments remains unclear. Methodological heterogeneity across participant’ characteristics, fatigue-induction protocols and criteria for its confirmation, and balance assessment tasks limits the generalizability of current findings. Future research employing sport-specific fatigue protocols and sensitive balance measures while accounting for relevant participant’s characteristics may help clarify the effects of fatigue on balance performance in athletes with CAI and inform more targeted rehabilitation and injury-prevention strategies.

Authors’ contributions

M.M.A.: Writing – review & editing, Writing – original draft, Project administration, Conceptualization, Investigation, Methodology, Validation, Visualization, Funding acquisition. R.M.: Writing – review & editing, Writing – original draft, Conceptualization, Investigation, Methodology, Validation, Visualization, Supervision, Data curation. N.N.A.: Writing – review & editing, Writing – original draft, Conceptualization, Methodology, Validation, Visualization.

Funding

The work was supported by the Ahvaz Jundishapur University of Medical Sciences (05s4).

Data availability

All data analyzed during this study are included in this published article and its supplementary information files. Additional data or materials related to this study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

IR.AJUMS.REC.1405.008.

Not applicable. This study is a systematic review of previously published studies.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.