Comparison of Lower Limb Muscle Strength Imbalances and Gait Performance in Postmenopausal Women with Healthy and Low Bone Mineral Density

Abstract

Introduction

Osteoporosis and osteopenia increase the risk of fractures, impair physical function, and may lead to loss of independence in postmenopausal women. This study examined differences in muscle strength imbalances and gait performance in postmenopausal women with healthy and low bone mineral density (BMD).

Methods

In this cross-sectional study, a total of 85 participants underwent body composition and BMD measurements using dual-energy X-ray absorptiometry (DXA, Hologic QDR Horizon A). Muscle strength was evaluated using a handgrip strength test, and knee flexor and extensor strength were analyzed using an isokinetic dynamometer (Humac Norm CSMI). Gait performance was evaluated by a 10 m walking test, a 3 m backward walking test, and a 6-minute walking test. According to whole-body BMD T-score, participants were divided into a Healthy BMD group (n = 30) and a Low BMD group (n = 55).

Normality was assessed using the Shapiro–Wilk test. Between-group differences and side-to-side differences in strength parameters were analyzed using the Mann–Whitney U-test, and associations between gait parameters, muscle strength, and BMD were examined using Spearman’s rank correlation coefficient. Additionally, multivariable analyses, including multivariate analysis of covariance (MANCOVA) and binary logistic regression, were performed to adjust for age and lean body mass. Statistical significance was set at p < 0.05.

Results

Low BMD status was associated with reduced muscle strength, particularly lower knee extensor strength. After adjustment for age and lean body mass, right knee extensor strength remained independently associated with low BMD status in binary logistic regression (OR = 0.957, 95% CI: 0.927–0.988, p = 0.007). No significant association was observed between gait performance and BMD.

Discussion

These results suggest that lower limb muscle strength, particularly knee extensor strength, may be a relevant functional factor to consider in postmenopausal women with low BMD, as reduced strength may be associated with increased fracture risk in older women.

Introduction

Osteoporosis and osteopenia are prevalent health concern among postmenopausal women and is characterized by low bone mineral density (BMD) and an increased risk of fractures.¹ Understanding the relationship between bone health and muscle strength is crucial for developing effective preventive and therapeutic strategies. The association between bone mineral density (BMD) and muscle strength in postmenopausal women is pivotal for understanding musculoskeletal health. Muscles and bones are interdependent, with muscle contractions exerting mechanical forces on bones that stimulate bone remodeling, while bones provide structural support for muscles.^2,3 Reduced muscle strength in postmenopausal women is correlated with decreased BMD, increasing the risk of osteoporosis and fractures.^4–6 Dual-energy X-ray absorptiometry (DXA) scans often reveal correlations between low BMD and diminished muscle mass, highlighting the importance of interventions such as resistance training and weight-bearing exercises to enhance both muscle strength and bone density.^7–9 Hormonal changes and nutritional factors also play significant roles, emphasizing the need for comprehensive strategies to preserve musculoskeletal health in aging women.^10,11

Kim and colleagues¹² investigated the connection between poor bone mineral density and weak grip strength. Their research indicated that low grip strength is linked to low bone mineral density in the hip, femur neck, and spine. These findings emphasize the role of overall muscle strength as a key indicator of bone health in aging populations. While grip strength is often used as a general proxy for musculoskeletal condition, lower limb muscle strength is even more directly tied to functional mobility and independence in postmenopausal women.

Lower limb muscle strength directly impacts mobility, balance, and independence, which can lead to difficulties in performing daily activities such as walking, climbing stairs, and getting up from a chair. This decline in lower limb muscle strength is often exacerbated in older women due to factors such as hormonal changes, decreased physical activity, and age-related conditions. Additionally, during a woman’s life, an imbalance in the strength of the lower limbs could develop.^13–15

Lower limb muscle strength asymmetry refers to an imbalance in strength between the muscles of one leg and those of the other. This imbalance can arise due to various factors such as injury, surgery, habitual movement patterns, or neurological conditions. In older women, asymmetry in lower limb muscle strength can be particularly concerning, as it may contribute to gait abnormalities, an increased risk of falls, and reduced functional mobility.^13,16 Addressing asymmetry typically involves targeted exercises to strengthen weaker muscles while maintaining strength in stronger muscles.^17–19 Lower limb muscle strength imbalances refer not only to differences between limbs (asymmetry), but also to disproportionate strength between opposing muscle groups within the same limb, such as the quadriceps and hamstrings. These imbalances can negatively impact joint stability, coordination, and functional performance, and are often associated with an increased risk of musculoskeletal injury and impaired mobility, particularly in older adults.²⁰

Research shows that measuring lower limb strength in older adults using both the right and left legs, rather than focusing exclusively on laterality (dominant versus non-dominant), is essential due to the changes in laterality that often occur as a result of disease, injury, or other age-related factors. This approach acknowledges that health conditions and injuries can affect strength dynamics differently across limbs, thereby providing a more comprehensive understanding of an individual’s functional capabilities.^17,21,22

The association between bone mineral density (BMD) and lower limb muscle strength imbalances in postmenopausal women underscores the intricate relationship between musculoskeletal health factors. Lower limb muscle strength imbalances can lead to altered loading patterns on bones during weight-bearing activities, potentially influencing bone remodeling processes and BMD. Hormonal changes associated with aging and menopause also play a role, affecting both muscle and bone health.²³ Therefore, this study aimed to investigate differences in lower limb strength imbalances and gait parameters in postmenopausal women classified into healthy and low BMD groups based on DXA-derived whole-body BMD T-score. In this study, the terms healthy BMD and low BMD refer to study-specific grouping based on whole-body BMD T-score and are used to examine musculoskeletal differences within the study sample.

Parts of cohort have been previously described in our earlier publication.²⁴ However, the published work focused on the effects of the intervention, whereas this study focused on examining differences in lower limb strength imbalances and gait and presents completely new ones that have not been presented before.

Methods

Participants

In this cross-sectional study, one hundred ten postmenopausal women aged over 60 years from an urban area in the Czech Republic initially agreed to participate. Eighty-five older women fulfilled the criteria for participation in the study. The recruitment period began on April 7, 2019, and ended on April 30, 2019. Participants were recruited to this study by the snowball method at recruitment locations such as the third-year university (an educational program designed for older adults, typically retirees, offering lifelong learning opportunities), libraries or various city offices. Moreover, public media, such as local newspapers, radio, and leaflets, were used for subject recruitment. This pre-pandemic recruitment period ensured that participants’ mobility, physical activity, gait performance, and general health status were not affected by COVID-19-related restrictions.

Inclusion criteria consisted of (1) ≥60 years old, (2) living independently, (3) passing the Montreal Cognitive Assessment (MoCA) with a score 26 and more, and (4) being able to walk ≥10 meters unaided. The exclusion criteria consisted of (1) severe musculoskeletal issues (eg., osteoarthritis, recent fractures), (2) neurological (eg., Parkinson’s, stroke) or oncological conditions, (3) the need for mobility aids, and (4) not being able to undergo a DXA examination.

According to the whole-body BMD T-score, the participants were divided into a healthy BMD group (n = 30) and a low BMD group (n = 55). A T-score threshold of –1.0 was used to distinguish participants with whole-body BMD values within the normal range from those with reduced whole-body BMD. Values higher than –1.0 indicated inclusion in the healthy BMD group, while values of –1.0 and below indicated inclusion in the low BMD group. In the present study, these group labels refer to study-specific stratification based on whole-body BMD T-score and should not be interpreted as a definitive clinical diagnosis of osteopenia or osteoporosis, as standard clinical DXA-based classification is typically based on central skeletal sites, such as the lumbar spine, total hip, and femoral neck. All participants provided written informed consent prior to study enrollment.

An a priori sample size calculation was performed using G*Power 3.1.9.7. The expected effect size was set at d = 0.5, with α = 0.05, power = 0.80, and a one-tailed test, resulting in a required total sample size of 106 participants. Although 110 postmenopausal women initially agreed to participate, only 85 fulfilled all eligibility criteria and completed the required assessments. Therefore, the final analytical sample was smaller than the calculated target sample size, which should be considered when interpreting the statistical power and robustness of the findings.

The study was conducted in accordance with the Declaration of Helsinki and approved under a broader research project by the Research Ethics Committee (REC) of Masaryk University, Brno, Czech Republic (EKV-2018-088-R1). One component of this project has been published previously;²⁴ the present manuscript reports findings from a separate component with a distinct research objective and previously unreported outcomes.

Procedures

The Montreal Cognitive Assessment (MoCA)

The MoCA is a 10-minute cognitive screening tool that has been validated as a highly sensitive tool for the early detection of mild cognitive impairment (MCI). The MoCA accurately assesses short-term memory, visuospatial abilities, multiple aspects of executive functions, attention, concentration, working memory, language and orientation to time and place. The total possible score is 30 points; a score of 26 or above is considered healthy.²⁵ All participants scored 26 or more points.

Anthropometry

A calibrated stadiometer (Seca 264) was used to measure height and weight. A tailor’s tape measure was used to determine waist and hip circumference. Based on these measurements, body mass index (BMI) and waist-to-hip ratio (WHR) were calculated.

DXA

Dual-energy X-ray absorptiometry (DXA; Hologic QDR Horizon A, Hologic Inc., Marlborough, MA, USA) was used to assess body composition and bone mineral density. DXA-derived variables included hip BMD, Ward’s region BMD, whole-body BMD, fat mass, and lean mass. All scans were performed by a certified operator according to standardized manufacturer-recommended procedures. Ward’s region BMD was included as an additional DXA-derived variable and was not used as the primary criterion for group allocation.

Muscle Strength Measurements

Knee flexor and extensor strength were analyzed using the isokinetic dynamometer Humac Norm CSMI (Stoughton MA, USA). Participants were seated in the isokinetic dynamometer chair with the back support adjusted to an 85° angle. The dynamometer pad was placed approximately 3 cm above the lateral malleolus. The testing protocol commenced with the evaluation of the dominant limb. Participants performed five nonmaximal trials on the dynamometer for each movement to warm up and become familiar with the motions. After thirty seconds of rest, concentric isokinetic knee flexion and extension movements were assessed at an angular velocity of 60 degrees per second (60°/s). Each movement included five maximal repetitions over a 90-degree range of motion, from 0° (full knee extension) to 90° of knee flexion. The maximal knee extensor and flexor strengths were determined by measuring the peak torque (in Newtons, Nm) during the isokinetic concentric contraction.

The maximal peak torque (in Newtons, Nm) of the knee extensors and flexors, normalized to body mass, and the hamstring-to-quadriceps ratio (H:Q) were further analyzed.

Muscle strength measurements and other physical assessments were conducted by trained members of the research team.

Hand Grip

The hand grip strength test measures the strength of the finger flexors (muscles of the forearm) using a hand dynamometer (TAKEI 5401 Dynamometer). The examinee sits upright on a chair without a backrest. The joints of the lower limb were positioned in approximately 90°–90°–90° triflexion (90°-90°-90°). The shoulders were in a neutral position, and the elbow of the examined arm was flexed at an angle of 90°. The opposite arm was placed loosely on the thigh (on the side of the examined arm). The strength of the hand grip was examined three times for each upper extremity with a rest interval of 2 minutes between tests. We tested the nondominant arm first and then the dominant arm. It is very important that the examinee does not know the interim results and is not continuously encouraged. The examiner should only give the following instructions, not evaluate the measured results in any way or share them with the subject. The best attempt was always included in the analysis.²⁶

Gait 10 m (10MWT)

Participants walked 10 meters unassisted, with the time recorded over the middle 6 meters to account for acceleration and deceleration phases. The test was performed at the highest possible speed in three trials. The best trial was selected for further analysis. A wireless Photocell (Brower timing systems) was used for measurement accuracy.²⁷

Gait Backward 3 m (3MBW)

A distance of 3 m was measured and marked with tape. The surface condition of the floor was wood. Participants were asked to align their heels with the tape, not allowed to break into a run during the test and were permitted to look behind themselves. The examiner walked backward with the participant to ensure safety. The participants completed three trials, and the average of all three trials was recorded.²⁸ A wireless Photocell device (Brower timing systems) was used for measurement accuracy.

6-min Walk Test (6MWD)

The goal of this test was to walk as fast as possible for six minutes along a marked route. The marked route was rectangular with a total length of 50 meters. Participants set their own pace and could stop to rest if they wished. After completing the test, the number of laps and extra meters traveled were counted.²⁹

Statistical Analysis

The Shapiro–Wilks test was used to test normality. As most of the variables did not meet the assumptions of normal distribution, the Mann‒Whitney U-test was used to compare differences between the healthy BMD group and the low BMD group. The Mann‒Whitney U-test was also used to analyze differences between the strength parameters of the right and left limbs. Spearman’s rank correlation coefficient was used to assess associations between gait parameters, muscle strength, and bone mineral density (BMD). The corresponding 95% confidence intervals for Spearman’s rho correlation coefficients are provided in S2 Confidence Intervals of Spearman’s rho (Supporting Information).

The a priori sample size calculation was performed using G*Power 3.1.9.7 and is reported in detail in the Participants section.

In addition to the primary non-parametric analyses, further multivariable analyses were performed to account for potential confounding factors. A multivariate analysis of covariance (MANCOVA) was conducted to examine differences between the healthy BMD and low BMD groups across multiple muscle strength outcomes simultaneously, while adjusting for age and lean body mass. Right-sided muscle strength variables, including handgrip strength, knee extensor strength, and knee flexor strength, were included as dependent variables, BMD group as a fixed factor, and age and lean body mass as covariates. Pillai’s trace was used to interpret multivariate effects due to its robustness to potential violations of covariance matrix homogeneity. When a significant multivariate effect was observed, follow-up univariate ANCOVA tests were performed. Partial eta squared (η²) was reported as a measure of effect size. Other potential confounders, including habitual physical activity, dietary intake, calcium/vitamin D supplementation, osteoporosis-related medication use, and comorbidities, were not available in sufficient detail for inclusion in the adjusted models.

Additionally, binary logistic regression analysis was performed to identify muscle strength variables independently associated with low BMD status. BMD group (healthy vs. low BMD) was used as the dependent variable. Muscle strength parameters were entered as independent variables in separate models to avoid multicollinearity, with age and lean body mass included as covariates. Odds ratios (ORs) with 95% confidence intervals (CIs) were calculated. Model fit was assessed using the Hosmer–Lemeshow goodness-of-fit test, and explanatory power was evaluated using Nagelkerke’s R².

All analyses were performed using IBM SPSS Statistics for Windows, Version 28.0 (IBM Corp., Armonk, NY, USA), and statistical significance was set at p < 0.05.

The dataset used for the analyses is provided as S1 Dataset (Supporting Information). The 95% confidence intervals for the reported Spearman’s rho correlation analyses are provided as S2 Confidence Intervals of Spearman’s rho (Supporting Information).

Results

The participants’ characteristics are shown in Table 1. No statistically significant differences were observed in age, height, body weight, BMI, WHR, fat mass, or lean mass between the healthy BMD group and the low BMD group.

Table 1.

Participant Characteristics

	Healthy BMD (n=30)		Low BMD (n=55)		p
	Mean	SD	Mean	SD
Age (years)	69,07	4,23	69,56	4,81	0,766
Height (cm)	162,76	5,50	160,62	5,28	0,051
Weight (kg)	75,97	17,71	71,52	11,92	0,478
BMI	28,66	6,44	27,76	4,59	0,859
WHR	0,93	0,09	0,91	0,06	0,612
Fat mass (kg)	30,95	11,07	29,80	72,33	0,945
Lean mass (kg)	45,32	66,85	42,79	65,67	0,071

Table 2 shows the results of the comparison between the healthy BMD and low BMD groups. The healthy BMD and low BMD groups significantly differed in all analyzed bone parameters (whole-body BMD, hip BMD, Ward’s BMD, p ≤ 0.001). Handgrip strength of the right and left upper limbs was significantly greater in the healthy BMD group (p = 0.018 and p = 0.004 for the right and left hands, respectively). The strength of the knee extensors (p<0.000 for both the right and left lower limbs) and the knee flexors of the right lower limb (p = 0.023) were significantly greater in the healthy BMD group than in the low BMD group. The H:Q ratio of the left lower limb was significantly greater in the low BMD group (p = 0.014). Gait parameters did not significantly differ between the analyzed groups.

Table 2.

Descriptive Statistics and Results of Group Comparisons for Muscle Strength, Muscle Function and Bone Parameters

	Healthy BMD (n=30)		Low BMD (n=55)		p
	Mean	SD	Mean	SD
Handgrip R (kg)	27,49	4,82	24,96	4,14	0,018*
Handgrip L (kg)	26,17	4,40	23,32	5,38	0,004*
10MWT (s)	3,20	0,84	3,17	0,77	0,823
3MBW (s)	2,77	0,85	2,83	0,71	0,561
6MWD (m)	565,17	76,93	556,91	59,11	0,316
Knee extensors R (kg)	98,40	16,97	84,11	19,32	0,000**
Knee flexors R (kg)	50,20	11,68	45,38	10,91	0,023*
H:Q ratio R	0,51	0,09	0,56	0,18	0,329
Knee extensors L (kg)	94,57	18,97	80,51	16,00	0,000**
Knee flexors L (kg)	48,30	12,79	45,35	11,33	0,278
H:Q ratio L	0,51	0,09	0,56	0,09	0,014*
BMD Ward’s (g/cm2)	0,75	0,26	0,59	0,22	0,001*
BMD Ward’s T score	−0,68	1,35	−2,21	0,68	0,000**
BMD Ward’s Z score	1,41	1,36	0,05	0,73	0,000**
BMD whole body (g/cm2)	1,10	0,07	0,94	0,05	0,000**
BMD whole body T score	−0,13	0,80	−2,15	0,71	0,000**

Notes: *Statistically significant at 0.05. **Statistically significant at 0.001.

When comparing handgrip strength between the right and left upper limbs, significantly greater values were measured for the right upper limb in both groups (p = 0.009 and p < 0.000 for the healthy BMD and low BMD groups, respectively). Differences in lower limb strength were statistically significant only in the low BMD group. Knee extensor strength significantly differed between the right and left limbs (p = 0.027). The extensors of the right lower limb were stronger in the low-BMD group. Additionally, the H:Q ratio was significantly greater for the left lower limb in this group (p = 0.045). Detailed values and statistical comparisons are presented in Table 3.

Table 3.

Differences in Strength Between the Right and Left Limbs

	Healthy BMD (n=30)	Low BMD (n=55)
	p	p
Handgrip (kg)	0,009*	0,000**
Knee extensors (kg)	0,071	0,027*
Knee flexors (kg)	0,060	0,781
H:Q ratio	0,734	0,045*

Notes: *Statistically significant at 0.05. **Statistically significant at 0.001.

Correlations between gait variables and strength parameters showed statistically significant associations (Table 4). The 10MWT correlated with handgrip strength in the healthy BMD group. In the low BMD group, a correlation was observed between the 10MWT and knee flexor strength as well as the H:Q ratio. The 6MWD correlated with handgrip strength, knee flexor strength, and the H:Q ratio in both groups. Additionally, a correlation between the 6MWT and knee extensor strength was observed in the low BMD group. Backward gait correlated with handgrip strength, knee extensor strength, and knee flexor strength in the healthy BMD group. In the low BMD group, a correlation between the backward gait and handgrip strength was observed. No statistically significant correlation between gait parameters and bone mineral density was observed. The 95% confidence intervals for the Spearman’s rho correlation coefficients reported in Table 4 are provided in S2 Confidence Intervals of Spearman’s rho (Supporting Information).

Table 4.

Correlation Between Gait Variables, Strength Parameters and Bone Mineral Density

	Healthy BMD (n=30)			Low BMD (n=55)
	10MWT	3MBW	6MWD	10MWT	3MBW	6MWD
	(s)	(s)	(m)	(s)	(s)	(m)
Handgrip R	−0.48*	−0.51*	0.63*	−0.26	−0.29*	0.27*
Handgrip L	−0.49*	−0.53*	0.57*	−0.13	−0.21	0.08
EXT_R	−0.25	−0.54*	0.29	0.01	−0.07	0.12
FLEX_R	−0.23	−0.49*	0.45*	−0.36*	−0.24	0.45*
HQ R	−0.12	−0.15	0.36	−0.34*	−0.21	0.31*
EXT_L	−0.27	−0.60*	0.36	−0.21	−0.21	0.31*
FLEX_L	−0.34	−0.47*	0.47*	−0.39*	−0.23	0.44*
HQ L	−0.31	−0.11	0.37*	−0.31*	−0.12	0.30*
BMD whole body (g/cm2)	0.00	0.20	−0.31	−0.08	0.05	−0.10
BMD whole body T score	0.01	0.20	−0.33	−0.03	0.05	−0.10

Notes: *Statistically significant at 0.05.

To account for potential confounding effects of age and lean body mass, additional multivariable analyses were performed. In the MANCOVA model (Table 5) including right-sided muscle strength variables, a significant multivariate effect of BMD group was observed after adjustment (Pillai’s trace = 0.108). Both age (p = 0.001) and lean body mass (p < 0.001) were significantly associated with the combined muscle strength outcomes.

Table 5.

MANCOVA Results for Muscle Strength Outcomes Adjusted for Age and Lean Body Mass

Outcome Variable	F	p-value	Partial η2
Multivariate effect (BMD group)	F(3,79) = 3.191	0.028	0.108
Handgrip R	3.643	0.060	0.043
EXT_R	8.608	0.004	0.096
FLEX_R	2.282	0.135	0.027

Notes: MANCOVA adjusted for age and lean body mass. Pillai’s trace was used for multivariate testing.

In the follow-up univariate analyses, right knee extensor strength remained significantly different between groups after adjustment (F = 8.608, p = 0.004, partial η² = 0.096), with higher values observed in the healthy BMD group. Handgrip strength showed a borderline association with BMD group (F = 3.643, p = 0.060, partial η² = 0.043), whereas right knee flexor strength was not significantly different between groups (F = 2.282, p = 0.135, partial η² = 0.027).

In addition, binary logistic regression analysis (Table 6) was performed to examine independent associations with low BMD status. In the model adjusted for age and lean body mass, right knee extensor strength was independently associated with low BMD status (OR = 0.957, 95% CI: 0.927–0.988, p = 0.007), indicating that higher knee extensor strength was associated with lower odds of low BMD. Age and lean body mass were not significant predictors in the model. The overall model was statistically significant (χ² = 11.94, p = 0.008), showed acceptable goodness-of-fit (Hosmer–Lemeshow p = 0.256), and explained approximately 18% of the variance (Nagelkerke R² = 0.180).

Table 6.

Binary Logistic Regression Analysis for Predictors of Low BMD

Variable	B	OR (Exp(B))	95% CI	p-value
EXT_R	−0.044	0.957	0.927–0.988	0.007
Age	−0.024	0.976	0.875–1.090	0.671
Lean body mass	0.000	1.000	1.000–1.000	0.534

Notes: Model fit: χ² = 11.94, p = 0.008; Nagelkerke R² = 0.180; Hosmer–Lemeshow p = 0.256.

Discussion

This study focused on differences in lower limb strength imbalances and gait in postmenopausal women with healthy and low BMD and provides pivotal insight into the complex interplay between muscle strength, bone health, and imbalances. The findings of this study highlight significant differences between the healthy BMD and low BMD groups. Significant differences were observed not only in bone parameters such as whole-body BMD and Ward’s BMD but also in lower limb strength balances, underscoring the importance of bone health in maintaining overall musculoskeletal integrity. The originality of this study lies in its focus on lower limb muscle strength imbalances rather than muscle strength alone in postmenopausal women with different BMD status. While previous studies have commonly examined associations between BMD, general muscle strength, physical performance, or gait parameters, less attention has been paid to side-to-side and agonist-antagonist strength imbalances as potential functional characteristics related to low BMD. By combining DXA-based BMD assessment, isokinetic evaluation of knee flexor and extensor strength, handgrip strength, and standardized gait performance tests, this study provides a more comprehensive view of musculoskeletal function in postmenopausal women.

The observed differences in handgrip strength and knee extensor/flexor strength between the two groups suggest an association between BMD status and muscle function or strength imbalances. Interestingly, gait parameters did not significantly differ between groups, suggesting that other factors may influence walking patterns in this population. There is disagreement among previous studies examining the relationship between walking and BMD. Some studies have suggested a positive association between brisk walking and BMD at various skeletal sites,^30,31 while others have shown conflicting results regarding the effectiveness of walking in improving BMD.³² Studies have also examined the relationship between walking speed and BMD, showing significant correlations between lower body strength, agility, and walking speed, and total bone mineral density and cortical thickness.³³ Although modern gait assessment methods using wearable inertial sensors or other motion-sensing technologies are increasingly used to monitor gait in real-life settings, we opted for standardized clinical walking tests performed in a controlled environment. These tests (10MWT, 3MBW, and 6MWD) are widely used in clinical research to assess functional mobility, gait speed, dynamic balance, and endurance. Combined with precise electronic timing systems, they provided a reliable and replicable framework for evaluating gait performance across participants. This approach ensured methodological consistency and measurement accuracy while reflecting clinically relevant walking behavior in older adults.

To further clarify whether these between-group differences were independent of potential confounding factors, additional multivariable analyses were performed. The MANCOVA results demonstrated that differences in muscle strength between groups persisted after adjustment for age and lean body mass, indicating that the observed group differences cannot be explained solely by these factors. However, follow-up analyses revealed that knee extensor strength was the only parameter that remained significantly different between groups after adjustment, whereas handgrip strength showed only a borderline association and knee flexor strength was no longer significant.

This finding was further supported by the logistic regression analysis, which showed that knee extensor strength was independently associated with BMD status even after adjusting for age and lean body mass. In contrast, neither age nor lean body mass remained significant predictors in the model. These results suggest that muscle function, particularly quadriceps strength, may be more strongly related to bone health than muscle mass alone. From a mechanistic perspective, this association is consistent with the role of knee extensors in generating mechanical loading during weight-bearing activities, which is a key stimulus for bone remodeling. However, this interpretation should be considered exploratory. Because the study was cross-sectional and several lifestyles, nutritional, pharmacological, and health-related confounders were not measured in detail, the results do not establish a causal relationship between knee extensor strength and BMD status. It is possible that reduced knee extensor strength reflects lower mechanical loading of the skeleton; however, this hypothesis cannot be confirmed from the present cross-sectional data.

For instance, Nonaka et al found no significant differences in gait speed, or the TUG test results among groups with varying BMD, in line with findings from other research that also did not observe correlations between these physical performance indicators and BMD in older women.³³ A justification for the lack of a statistically significant correlation may lie in the inherent biological complexity of older women’s physiology, where changes in BMD might be more profoundly influenced by hormonal changes, muscle mass alterations, and psychomotor coordination rather than gait performance alone. This perspective is supported by studies recognizing multifactorial influences as critical in determining both gait and bone density outcomes.^34,35

Lower limb strength imbalances, particularly in the low BMD group, raise questions about mobility and fall risk implications in older women. Previous studies^36–38 investigated the relationship between muscle strength imbalances, particularly in the knee extensors and flexors, and their implications for mobility and fall risk in older women. LaRoche et al³⁶ investigated how strength asymmetry in the knee extensors influences gait asymmetry and variability in older women, highlighting the impact of muscle imbalances on functional abilities and balance, which are crucial factors in determining fall risk. Perry et al³⁹ and Skelton et al⁴⁰ reported greater asymmetry in leg extension power and lower limb explosive power in older fallers compared to nonfallers, highlighting the importance of assessing symmetry in strength and power to understand fall risk. The observed muscle strength imbalances, particularly in the low BMD group, may be explained by the well-documented relationship between bone and muscle health. The musculoskeletal system functions as an integrated unit in which bone and muscle are interconnected through mechanical and biochemical pathways. Although the coexistence of low bone mineral density and reduced muscle mass or strength is often discussed in the context of osteosarcopenia, osteosarcopenia was not directly assessed in the present study. Therefore, our findings should be interpreted as associations between BMD status and selected measures of muscle strength and strength imbalance, rather than as evidence of osteosarcopenia.

The greater handgrip strength and lower limb strength in the healthy BMD group may indicate a potential protective effect of muscle strength against bone loss. Preceding studies^40,41 have provided relevant insights into the relationship between muscle strength and bone health in women aged 60 and above. These studies indicate that greater muscle strength, particularly in the lower extremities, is associated with better bone mineral density, indicating a potential protective effect of muscle strength against bone loss in this population. As shown in recent intervention research, even moderate physical activity programs can significantly improve lower limb strength and functional performance in postmenopausal women, underscoring the potential for non-pharmacological strategies to maintain mobility and quality of life⁴¹ These findings suggest that maintaining or improving muscle strength, especially in older women, may play a crucial role in preserving bone health and reducing the risk of osteoporosis.

The identification of statistically significant differences in handgrip and lower limb muscle strength imbalances in the low BMD group underscores the need for further research to analyze the underlying mechanisms and to create targeted interventions to preserve musculoskeletal health and reduce fracture risk in this population.

Taken together, the results of both multivariable approaches consistently identify knee extensor strength as the most robust muscle-related factor associated with low BMD in this population.

This study has several limitations. First, the cross-sectional design precludes causal inference. Therefore, the observed differences and associations between BMD status, muscle strength imbalance, and gait performance should be interpreted cautiously, and the directionality of these relationships cannot be determined. Furthermore, because participant allocation was based on whole-body BMD T-score rather than standard central DXA diagnostic sites, the healthy BMD and low BMD groups should be interpreted as study-specific categories rather than definitive clinical diagnostic groups, which may limit the direct clinical interpretability of the findings. Second, the final sample size was lower than the a priori calculated sample size, which may have limited the statistical power to detect smaller between-group differences or weaker associations. Moreover, the study included only 85 participants, which may limit the generalizability of the findings. A larger, more diverse sample could provide more robust results and increase the external validity of the study. The participants were recruited from a single city in the Czech Republic, which might limit the applicability of the findings to other populations with different socioeconomic and healthcare contexts. In addition, data were collected in 2019, and the time elapsed between data collection and publication should be considered when interpreting the findings. Because the measurements are based on objective DXA, isokinetic strength, and standardized gait assessments, the dataset remains suitable for addressing the cross-sectional research question, although the time elapsed is acknowledged as a limitation. However, the pre-pandemic recruitment period also indicates that the measurements were not influenced by COVID-19-related restrictions affecting mobility, physical activity, or general health status.

Future longitudinal studies are needed to determine the temporal and potentially causal relationships between bone density, muscle strength imbalance, and gait performance. Although the inclusion criteria excluded selected neurological, oncological, and severe musculoskeletal conditions, several potentially important confounders were not systematically controlled for. Detailed information on habitual physical activity, dietary intake, calcium/vitamin D supplementation, osteoporosis-related medication uses such as bisphosphonates, and selected metabolic or inflammatory comorbidities were not available in sufficient detail for statistical adjustment. These factors may influence both BMD and muscle strength and could therefore have affected the observed associations. Accordingly, the findings should be interpreted cautiously and as associations rather than causal relationships. Future studies should collect these variables prospectively and include them in adequately powered multivariable models.

Although this study focused on specific muscle groups, namely handgrip strength and knee flexors and extensors, we acknowledge that it does not represent a full assessment of global muscle strength imbalances. Previous research has demonstrated that grip strength correlates well with overall muscle strength in older adults, supporting its use as a proxy in geriatric populations. Nonetheless, future studies should consider including a wider range of muscle groups and bilateral assessments to better capture global patterns of muscle strength imbalances and provide a more comprehensive view of musculoskeletal aging.

Conclusions

In conclusion, our investigation of lower limb muscle strength, its imbalances, and gait performance in postmenopausal women with healthy and low BMD provides additional insight into the complex interplay between musculoskeletal factors. The significant differences observed in muscle strength parameters, particularly knee extensor strength, suggest that lower limb strength may be associated with BMD status in postmenopausal women, although causal conclusions cannot be drawn.

From a clinical perspective, these findings indicate that muscle strength imbalance assessment may be a useful addition to routine evaluation of postmenopausal women with low BMD. Such assessments could help identify individuals who may benefit from targeted strength training, fall-prevention strategies, and individualized rehabilitation aimed at preserving mobility and reducing fracture risk. Interestingly, despite clear differences in strength and BMD, gait parameters did not significantly differ between groups, suggesting that walking performance may be influenced by additional factors beyond bone and muscle health. Future longitudinal studies are recommended to explore causal relationships and the effectiveness of specific rehabilitation strategies.

Funding Statement

This study was supported by funding from Masaryk University (ID=49011).

Ethics Approval and Consent to Participate

All participants provided written informed consent before entering the study. The procedures performed in this study were anonymous and conducted in accordance with the Declaration of Helsinki. The study was reviewed and approved by the Research Ethics Committee (REC) of Masaryk University, Brno, Czech Republic (Approval ID: EKV-2018-088-R1).

Disclosure

The authors report no conflicts of interest in this work.