THE RELATIONSHIP BETWEEN SINGLE LEG BALANCE AND ISOMETRIC ANKLE AND HIP STRENGTH IN A HEALTHY POPULATION

ABSTRACT

Background:

Impaired balance and strength commonly affect athletes with conditions like chronic ankle instability (CAI). Yet, clinical research surrounding the relationship between balance, strength, and CAI is still growing. Deeper investigation of these relationships is warranted to better inform clinical practice patterns when managing athletes with balance deficits.

Purpose:

To investigate the relationship between single leg balance, ankle strength, and hip strength in healthy, active adults.

Study Design:

Observational study

Methods:

Forty healthy participants (age 23.7 ± 4.9 years) were assessed for static balance, using a modified version of the Balance Error Scoring System (mBESS), as well as isometric strength of ankle and hip musculature via handheld dynamometry. Pearson's correlations were used to analyze relationships between balance and strength measures. Paired t-tests were utilized to compare dominant and non-dominant limb performance.

Results:

Negligible to low, negative correlations were found between balance scores and hip extension strength (r = -0.24 to -0.38, p<0.05). High, positive correlations were found between ankle and hip strength measures (r = 0.75 to 0.84, p<0.05). When comparing dominant to non-dominant limbs, only minimal differences were noted in ankle eversion strength (mean difference = 6.0%, p<0.01) and hip extension strength (mean difference = 5.5%, p<0.01).

Conclusions:

Minimal relationships were identified between static balance and isometric ankle and hip strength. Comparison of dominant and non-dominant limbs suggests that clinicians should expect relative symmetry in balance and strength in healthy adults. Thus, asymmetries found during clinical examination should raise suspicion of specific impairments that may lead to dysfunction.

Level of Evidence:

2c

INTRODUCTION

In high school sports, ankle sprains account for 16% of all injuries, at a rate of 3.65 sprains per 10,000 athlete exposures.¹ Residual symptoms following ankle sprains include pain, instability, swelling, and recurrent sprains, leading to chronic ankle instability (CAI).² In fact, high school athletes with prior ankle injuries are 4.22 times more likely to suffer a second ankle injury, compared to those without a history of injuries.³ Several risk factors have been hypothesized to contribute to the development of CAI, including impairments of balance,^4-6 neuromuscular control,^4,7,8 hip strength,^8,9 ankle strength,^4-7,10-12 and ankle range of motion.⁵

Balance is a sensorimotor process aimed at maintaining, achieving, or restoring a state of stability during activity.¹³ It involves synchrony between proprioceptive, somatosensory, vestibular, visual, and neuromuscular systems. An athlete's ability to maintain balance on one leg, known as single leg balance (SLB), has been linked to ankle injuries.^14,15 McGuine et al¹⁴ found that high school basketball players who demonstrated poor SLB were 6.7 times more likely to sustain ankle injury than those with good balance. Similarly, Trojian et al¹⁵ found that athletes with poor SLB who did not receive in-season ankle taping were 8.82 times more likely to sustain an ankle injury. Therefore, assessing SLB can provide meaningful information about injury risk.

Maintaining SLB requires neuromuscular demands from the trunk, hip, knee, ankle and foot. Consequently, understanding these relationships can be challenging. The relationship between balance, ankle strength, and injury continues to be a conundrum in clinical research. It is hypothesized that neuromuscular control of ankle eversion and inversion musculature plays a role in ankle stability during sports.^4,6 Researchers have confirmed deficits in ankle eversion strength^10,16 and inversion strength¹⁰ in individuals with chronic ankle instability (CAI), but Kaminski et al¹¹ reported no differences. In a meta-analysis, Witchalls et al⁶ identified that weaker eccentric inversion strength was a risk factor for ankle injuries. However, no researchers have assessed the direct relationship between clinical measures of balance and ankle strength.

While the impact of hip strength on ankle injury risk is unclear,^8,9 some researchers have reported a connection between balance and hip strength.^17,18 Lee and Powers¹⁸ reported that individuals with weak hip abductors had increased peroneus longus activation during both static and dynamic balance tasks (d = 0.77). Also, individuals with CAI were found to have worse dynamic balance, hip external rotation strength, and hip abduction strength when compared to healthy peers.¹⁹ Yet, the authors are not aware of any studies that have assessed a direct relationship between static balance and hip strength.

Despite the high prevalence and physical impact of CAI, scientific literature does not conclusively define a relationship between specific impairments and recurrent injury risk. Perhaps part of the problem is the limited understanding of the actual relationships between balance and strength. This study aims to investigate the relationship between single leg balance, ankle strength, and hip strength in healthy, active adults.

METHODS

Design

The study protocol was approved by the University of South Dakota Institutional Review Board. All participants signed an approved informed consent form prior to participation. The study utilized a correlational study design comparing balance to ankle and hip strength in healthy, active adults.

Participants

All participants were recruited from a sample of convenience at a Midwestern university. Inclusion criteria were: age between 18-64 years, English-speaking, and a score of 5 or greater on the Tegner Activity Scale (indicating a minimum activity level of jogging twice weekly).²⁰ Exclusion criteria were: pregnancy, known balance impairment (unresolved head injuries, vertigo, recent head cold, etc), or a lower extremity injury in the prior six months.

Procedures

Demographic information was collected, including age, height, weight, gender, dominant leg, activity level, and lower extremity injury history. Participants completed a five-minute warm-up on a stationary bicycle at a light to moderate intensity. Next, participants were asked to remove footwear and socks, and randomly assigned to one of three stations: balance, ankle strength, and hip strength. Once testing was completed at each station, participants proceeded to subsequent stations in stepwise fashion. Examiners at each station provided standardized verbal instructions. At each station, one examiner was responsible for testing all participants in order to reduce errors from examiner agreement. Examiners were blinded from the participants’ results in other stations. Intra-rater reliability was calculated for each test using intraclass correlation coefficients (ICC) with 95% confidence intervals, based on a single rater, single measurement, absolute agreement, two-way mixed-effects model. Strength of reliability was defined as: < 0.5 (poor), 0.5-0.75 (moderate), 0.75-0.90 (good), and 0.90-1.0 (excellent).^21,22

Modified Balance Error Scoring System (mBESS)

The original BESS, which also includes double limb and tandem stances, has been reported with moderate to excellent intra-rater reliability (ICC = 0.60-0.92), moderate to good inter-rater reliability (ICC = 0.57-0.85), and moderate test-retest reliability (ICC_2,1 = 0.70).²³ Individual single leg balance scores have been reported with moderate to excellent intra-rater reliability (ICC = 0.50-0.95)^24,25 and moderate to good inter-rater reliability (ICC = 0.53-0.83).²⁴ Docherty et al²⁶ found that individuals with CAI had worse scores on the specific single leg balance items. Therefore, a modified version of the test (mBESS), consisting of only the single leg balance items, was used in this study.

The mBESS consisted of a single leg balance test on stable and unstable surfaces (Figure 1). First, participants stood on one leg, placed hands on their hips, and were asked to maintain balance with eyes closed for 20 seconds. Next, participants repeated the task on a foam pad (6 cm thickness, Airex AG, Switzerland). The examiner provided verbal and visual instruction and scored the tests in real time. Each of the following errors represented one point: opening eyes, lifting hands off hips, stepping, stumbling or falling out of position, lifting forefoot or heel, abducting the hip by more than 30°, or failing to return to the test position in more than five seconds.^23,26 Participants were allowed one practice trial, followed by two scored trials for each task. Right and left limbs were tested in random order. A 10-second rest period was provided between trials. For each limb, the average of the two trials was scored under the stable (ground) and unstable (foam pad) conditions. The mBESS score for each limb equaled the sum of stable and unstable scores. Intra-rater reliability was moderate for stable (ICC = 0.54; 95% CI: 0.37-0.68), unstable (ICC = 0.73; 95% CI: 0.61-0.82), and composite mBESS scores (ICC = 0.71; 95% CI: 0.58-0.80).

Figure 1.

Modified BESS testing on (a) stable and (b) unstable surface.

Strength Measures

For all strength measures, examiners utilized a digital hand-held dynamometer (HHD; MicroFET 2, Hoggan Scientific LLC, Salt Lake City, UT) to measure torque (Nm). Examiners marked locations, specified below, to consistently apply a downward force for five seconds or until the participants could not sustain the hold any longer. Right and left limbs were tested in random order. Three successful trials were recorded on each limb, with a 10-second rest period between tests. The average of three trials for each limb was converted to a torque-to-body weight (T:BW) value.

Isometric Ankle Strength

Isometric ankle strength was assessed using modified MMT techniques (Figure 2).²⁷ For ankle inversion, participants were positioned on their side with the ankle in a resting position (approximately 10-20° of ankle plantarflexion, neutral inversion-eversion). The examiner marked on the medial aspect of the foot, 5 cm proximal to the head of the first metatarsal, to be used as the point of applied force during the MMT. The examiner measured the distance between this point and the ankle joint axis to record the moment arm. Next, the examiner placed the HHD on the mark and applied a downward force to obtain the strength measurement.

Figure 2.

Strength testing of ankle (a) eversion and (b) ­inversion.

For ankle eversion strength, participants were positioned on their opposite side with the ankle in a resting position. The examiner marked on the lateral aspect of the foot, 2 cm proximal to the head of the fifth metatarsal, to be used as the point of applied force during the MMT. The examiner measured the distance between this point and the ankle joint axis to record the moment arm. Next, the examiner placed the HHD on the mark and applied a downward force to obtain the strength measurement. Intra-rater reliability was good for ankle inversion (ICC = 0.83; 95% CI 0.76-0.88) and eversion (ICC = 0.85; 95% CI 0.79-0.89).

Isometric Hip Strength

Isometric hip strength was assessed using standard manual muscle testing (MMT) procedures for hip extension and hip abduction (Figure 3).²⁷ For hip extension strength, participants were positioned in prone with their pelvis stabilized to the table using a belt. The examiner marked 10 cm proximal to the knee joint axis on the posterior thigh, to be used as the point of applied force during the MMT, then measured the distance between this point and the hip joint axis to record the moment arm. Next, participants flexed their knee to 90° and extended the hip of the tested limb. The examiner identified a mid-range of hip extension, placed the HHD on the mark, and applied a downward force to obtain the strength measurement.

Figure 3.

Strength testing of hip (a) extension and (b) abduction.

For hip abduction strength, participants were positioned on their side with their pelvis stabilized to the table using a belt. The examiner marked 5 cm proximal to the lateral femoral condyle, to be used as the point of applied force during the MMT. The examiner measured the distance between this point and the hip joint axis to record the moment arm. The participants abducted their tested hip with the knee extended. The examiner identified a mid-range of hip abduction, placed the HHD on the mark, and applied a downward force to obtain the strength measurement. Intra-rater reliability was excellent for hip extension (ICC = 0.93; 95% CI 0.90-0.95) and abduction (ICC = 0.91; 95% CI 0.87-0.94).

Data Analysis

Results from the mBESS composite score, stable and unstable balance scores, and isometric ankle and hip strength were analyzed using SPSS Statistics 25.0 Software (IBM, NY, USA). A significance of α = 0.05 was set for all analyses. Pearson's correlation (r) tests were utilized to analyze the relationship between balance and strength measures, as well as between individual strength measures. Strength of correlations were defined as: < 0.30 (negligible), 0.30-0.50 (low), 0.50-0.70 (moderate), 0.70-0.90 (high), and 0.90-1.0 (very high).²⁸ Paired t-tests compared balance and strength results between dominant and non-dominant limbs. The effect size of limb dominance was calculated by Cohen's d. An a priori power analysis revealed that 34 participants were needed to achieve an 80% study power with α error probability of 0.05 and effect size of d = 0.50.²⁹

RESULTS

Forty participants (age: 23.7 ± 4.9 years; 55% female; Tegner score: 6.9 ± 1.5) were included in data analysis. Descriptive statistics are highlighted in Table 1. Participants averaged 2.23 ± 1.57 errors per limb while balancing on a stable surface and 6.95 ± 1.77 errors per limb on an unstable surface, translating to an average mBESS score of 9.18 ± 2.90 errors per limb. Regarding ankle strength, participants averaged 0.33 ± 0.05 Nm/kg and 0.39 ± 0.05 Nm/kg of inversion and eversion strength, respectively. For hip strength, participants averaged 1.25 ± 0.34 Nm/kg and 1.65 ± 0.29 Nm/kg of extension and abduction strength, respectively.

Table 1.

Descriptive statistics (n=40).

	Mean (SD)
Age (years)	23.7 ± 4.9
Gender (% female)	55%
BMI	24.0 ± 3.3
Tegner Score	6.9 ± 1.5
mBESS Score (Errors)	9.18 ± 2.90
SLB Stable (Errors)	2.23 ± 1.57
SLB Unstable (Errors)	6.95 ± 1.77
Hip Extension (Nm/kg)	1.25 ± 0.34
Hip Abduction (Nm/kg)	1.65 ± 0.29
Ankle Inversion (Nm/kg)	0.33 ± 0.05
Ankle Eversion (Nm/kg)	0.39 ± 0.05
Eversion:Inversion Ratio	1.19 ± 0.16

BMI = body mass index, mBESS=Modified Balance Error Scoring System, SLB = single limb balance

Table 2 lists the results of the correlation analysis between balance and strength measures. There were negligible to low, negative correlations between balance tests and hip extension strength (r = −0.24 to −0.38, p<0.05). Higher scores (errors) on balance tests were associated with lower hip extension strength. That is to say, participants with worse single leg balance were more likely to have weaker hip extension strength. No other significant correlations between balance and strength were observed.

Table 2.

Pearson correlation (r) results between balance testing and torque to body weight strength measures (Nm/kg)

	Hip Extension	Hip Abduction	Ankle Inversion	Ankle Eversion	Ev:Inv Ratio
mBESS	-0.36†	-0.10	-0.07	-0.19	-0.08
SLB Stable	-0.24*	-0.83	-0.39	-0.13	-0.06
SLB Unstable	-0.38†	-0.10	-0.08	-0.19	-0.09

* Indicates a significant correlation (p<0.05); † Indicates a significant correlation (p<0.01).

Correlation matrices between ankle and hip strength are displayed in Table 3. There were low to moderate correlations between all T:BW values (r = 0.32 to 0.53, p<0.01) and high correlations between all torque values without bodyweight adjustment (r = 0.75 to 0.84, p<0.01).

Table 3a.

Pearson correlation matrix (r) between torque to body weight values (Nm/kg) of hip and ankle

	Hip extension	Hip abduction	Ankle inversion	Ankle eversion
Hip extension	1	0.41†	0.32†	0.46†
Hip abduction		1	0.47†	0.43†
Ankle inversion			1	0.53†
Ankle eversion				1

** Indicates a significant correlation (p < .01).

Comparison of dominant and non-dominant limbs are summarized in Table 4. Limb dominance had medium effects on hip extension and ankle eversion strength (d = 0.75 and 0.77, respectively), and a large effect on ankle eversion:inversion (E:I) ratio (d = 0.86). The dominant limb tested 5.5% stronger than the non-dominant limb for hip extension (mean difference: 0.07 Nm/kg, 95% CI 0.02-0.12) and 6.0% stronger for ankle eversion (mean difference: 0.02 Nm/kg, 95% CI 0.01-0.04). The E:I ratio was also higher on the dominant limb (p<0.01). No other measures demonstrated a significant difference in limb dominance.

Table 3b.

Pearson correlation matrix (r) between torque values (Nm) of hip and ankle strength tests.

	Hip extension	Hip abduction	Ankle inversion	Ankle eversion
Hip extension	1	0.75†	0.75†	0.82†
Hip abduction		1	0.79†	0.78†
Ankle inversion			1	0.84†
Ankle eversion				1

† Indicates a significant correlation (p < .01).

Table 4.

Statistical comparisons of test scores between dominant and non-dominant limb. Strength value differences are expressed as torque to body weight (Nm/kg)

	Percent Change	Mean Difference	95% Confidence Interval		T-score	Cohen'sd
			Lower	Upper
mBESS	NA	-0.01	-0.60	0.57	-0.43	-0.10
SLB Stable	NA	0.16	-0.23	0.55	0.84	0.19
SLB Unstable	NA	-0.18	-0.59	0.24	-0.86	-0.19
Hip Extension	5.5%	0.07†	0.02	0.12	3.35	0.75
Hip Abduction	1.8%	-0.03	-0.09	0.03	-0.14	-0.03
Ankle Inversion	3.3%	-0.01	-0.02	0.00	-1.73	-0.39
Ankle Eversion	6.0%	0.02†	0.01	0.04	3.43	0.77
E:I Ratio	NA	0.12†	0.06	0.18	3.83	0.86

mBESS = modified balance error scoring system; SLB = single leg balance; E:I = Eversion:Inversion † Indicates a significant difference (p<.01).

DISCUSSION

The results in this study suggest that minimal relationship exists between static balance and isometric strength of ankle and hip musculature in healthy, active adults. Only negligible to low correlations between balance and hip extension strength were observed, while all other correlations were insignificant. Two other investigators have also examined the balance-hip strength relationship, though both used the Y-Balance Test (YBT), a measure of dynamic balance.^30,31 Lee et al³⁰ found that in women, high correlations existed between YBT reach distances and isometric strength of hip extensors (r = 0.70 to 0.75) and abductors (r = 0.72). Culiver et al³¹ also identified an inverse correlation between hip abduction strength asymmetry and dominant limb YBT composite score (r = -0.46) in Division I baseball pitchers. These higher correlations may reflect the greater demands on muscle strength during a dynamic task (YBT), compared to the static single leg balance tasks examined in this study.

Nonetheless, the primary results mirror the conclusions made by a majority of researchers on healthy individuals.^32-35 In a meta-analysis, Muehlbauer et al³² concluded that little to no correlations existed between balance and lower extremity strength in children (r = 0.11, 95% CI −0.18 to 0.41), young adults (r = 0.20, 95% CI −0.02 to 0.41), or older adults (r = 0.27, 95% CI 0.15 to 0.40). Lin et al³³ and Dabadghav³⁴ found no correlation between static balance and strength of ankle evertors or invertors, as well as E:I ratio. Thus, in healthy individuals, successful maintenance of static balance does not rely solely on a single measure of ankle or hip strength.

Yet, individuals with ankle instability commonly have deficits in balance^19,36,37 and neuromotor function of the ankle^{7,10,11,36,38} and hip.^19,39 Several researchers have addressed interventional strategies for these deficits, including balance training,^40-45 ankle^44-47 and hip⁴⁸ strengthening, joint mobilization,^42,49 and external support.^50-53 Interestingly, some researchers have identified a direct effect of balance training on strength, and vice versa.^48,54,55 For example, Cug et al⁵⁴ found that four weeks of balance training using a BOSU® ball improved ankle strength in all directions. Conversely, Son et al⁵⁵ found that four weeks of isokinetic ankle strengthening improved single leg balance. Additionally, Smith et al⁴⁸ found that four weeks of supervised resistance band exercises, targeting hip abductor and external rotator strength, improved both static and dynamic balance in individuals with CAI.

One explanation for the differences between healthy and ankle instability populations may lie in the identification of thresholds. An open chain, maximal effort strength test may not accurately reflect the biomechanical demands of maintaining a closed chain, sub-maximal effort, single leg balance position. Hence, maintaining balance may rely on a minimum threshold of strength, rather than maximum strength. This could better explain the close relationship between impaired strength and balance following injury, in which pathological muscle weakness may fall below a threshold that affects balance. Future research could explore these thresholds, as well as muscle rate of force development (RFD). Additionally, authors have investigated muscle activation patterns during functional movements.^56-58 Overall, the larger body of literature still reinforces the idea that balance is a complex, multifactorial task.

Secondary Findings

In the current study, there were low to moderate correlations in strength normalized to body weight (T:BW), but high correlations in pure strength (torque). To the authors’ knowledge, this is the first study to report this peculiar pattern. While a sound biomechanical explanation requires a deeper investigation beyond the scope of this study, the results have important clinical implications. First, clinicians may save time by calculating strength as a torque value, rather than normalizing torque to body weight. Second, clinicians should expect that if an individual tests strongly in one muscle group, then they should test proportionately strong in other areas around the ankle and hip. Any disproportionate weakness may indicate an abnormal finding.

Regarding limb dominance, no asymmetries were detected in mBESS or static balance on stable and unstable surfaces. These results are met with mixed conclusions by other investigators.^33,34,59,60 Alonso et al⁵⁹ found that males aged 20-40 years had no limb asymmetries in balance, which was tested on a Biodex Balance System. In healthy adults (females and males), Lin et al³³ found no limb asymmetries in balance using a force platform. In contrast, Dabadghav³⁴ found that basketball players had lower radial displacement while balancing on their dominant limb, as tested on a force platform. Additionally, Promsri et al⁶⁰ reported that young, active adults prioritized different motor control strategies between dominant and non-dominant limbs, using a 3-dimensional motion capture system. However, they found no differences in the overall coordinative structure of balance. These conflicting results may be due to differences in population, testing procedures, or data analysis. First, basketball players may represent a unique population due to the postural demands of the sport. Secondly, researchers used varying methods to determine the “dominant limb.” Thirdly, balance testing varied between studies, ranging from a 10-second test under stable conditions³⁴ to a 20-second eyes closed test on an unstable surface. Finally, robust data capturing systems were able to identify more specific differences in balance (i.e. motor control).^34,60 Yet, the clinical relevance of this detail must be further investigated. In general, limb asymmetries should not be expected in healthy individuals when clinically testing SLB.^33,59

In this study, the dominant limb tested slightly stronger than the non-dominant limb for ankle eversion (6.0%), but not inversion. This corresponded to a significant increase in E:I ratio. This conflicts with conclusions by Dabadghav³⁴ and Lin et al,³³ who did not detect asymmetries in eversion strength, inversion strength, or E:I ratios. The small differences in this study may be related to testing methods. This study used isometric strength tests, whereas the two other authors utilized isokinetic tests. This increased the overall E:I ratios to 1.19 ± 0.16, which was higher than other reported ratios (0.80-1.06) in healthy individuals.^33,34 Interestingly, Lin et al³³ identified a significant pattern that slower isokinetic testing speeds resulted in higher E:I ratios. This might explain why isometric testing shifted the ratios in this study toward eversion. Despite differences, the overall relationship between limb dominance and ankle strength is minimal at best.

In the current study, hip extension strength tested slightly stronger in the dominant limb (5.5%), while hip abduction strength did not reveal any asymmetries. Lopes et al⁶¹ reported that healthy, male Navy cadets had slightly stronger dominant limb hip external rotation strength by 7.7% (p ≤ 0.05), whereas female cadets did not. Neither gender had differences in hip abduction strength. It is possible that the gluteus maximus, which is responsible for both hip extension and external rotation, may reveal limb dominance patterns, albeit minimal. However, the gluteus medius does not appear to be affected by limb dominance in healthy adults.

In summary, limb asymmetry during static balance and strength testing (generally > 6%) should not be explained by limb dominance. Furthermore, ankle and hip muscles should test proportionately strong. It is plausible for clinicians to interpret asymmetrical balance or strength, and disproportionate ankle or hip strength, as abnormal. Finally, the minimal relationship between balance and strength measures serves as a reminder of the complex, multifactorial nature of maintaining balance. Thus, treatment of balance and strength impairments should be individually tailored and complimentary in nature.

Limitations

The study collected data from a relatively small sample of healthy, active adults, thus limiting generalizability to a wider population. It is possible that test selection may have limited the results. For example, the mBESS may have both floor and ceiling effects which require further investigation. Despite this, the mBESS was selected because of its ability to efficiently quantify balance impairments with minimal equipment expense. Finally, the observational nature of this study should preclude the claim of any causal relationships.

CONCLUSION

In healthy, active adults, there is minimal relationship between static balance and isometric ankle or hip strength. However, high correlations exist between all ankle and hip strength measures. Furthermore, limb dominance does not seem to significantly impact performance on most tests, and at most explains a 6% asymmetry in strength. Therefore, in individuals with anticipated balance deficits, such as those with CAI, it is important for clinicians to assess and treat each impairment independently, using disproportionate weakness and/or asymmetry as a valid finding.