Relationships between walking speed, T-score and age with gait parameters in older post-menopausal women with low bone mineral density

Highlights

• Speed explained most of the variance in temporal-spatial and kinematic variables.

• T-score and age explained changes in some parameters related to dynamic stability.

• T-score increased the explained variance for hip and knee joint kinematics.

• Fast walking could maintain BMD by increasing mechanical loading on the bones.

Abstract

Background

The gait patterns of women with low bone mineral density (BMD) or osteoporosis have not been thoroughly explored, and when examined, often studied in relation to falls and kyphosis.

Research question: The aim of this study was to investigate the relationships between gait parameters and comfortable, self-selected walking speed and BMD in older post-menopausal women with a broad range of T-scores (healthy to osteoporotic).

Methods

3D kinematic and kinetic data were collected from forty-five women mean (SD) age 67.3 (1.4) years during level walking at their preferred speed. Multiple regression analyses explored the explained variance attributable to speed, femoral neck T-score, and age.

Results

The mean (SD) walking speed 1.40 (0.19) m·s⁻¹ explained the variance in most temporal-spatial, kinematic and joint powers (R² = 12–68%, P ≤ 0.01). T-score accounted for (R² = 23%, P ≤ 0.001) of the shared explained variance in stride width. It also increased the explanatory power for knee flexion (R² = 7%, P ≤ 0.05) and knee range of motion (R² = 12%, P ≤ 0.01). Power absorption by the knee flexors in terminal swing (K4) was the only power burst resulting in significant slope coefficients for all predictor variables (R² = 52 and 54%) (P ≤ 0.001) and (R² = 68%, P ≤ 0.05).

Significance

Speed alone explained most of the variance in the gait parameters, while speed and T-score combined increased the explanatory power of the regression models for some of the knee joint variables. Our findings demonstrated that older post-menopausal women, with a broad range of T-scores, are able to walk at comfortably fast speeds, generating gait patterns similar to those of younger women. The results also suggest that strengthening the hip abductor, knee extensor and flexor muscle groups may benefit the gait patterns of older postmenopausal women with low BMD.

Introduction

Osteoporosis is a systemic skeletal disease that is identified by decreased areal bone mineral density (BMD, g/cm²) and bone loss [1]. Osteoporosis can be a debilitating condition that predisposes primarily older women to an increased risk of fractures [1]. Every year, more than 8.9 million fractures worldwide are associated with osteoporosis, with 70% and 80% of these fractures occurring at the hip and forearm of women between 50–75 years of age, respectively [2]. The consequences of a fracture can be devastating with detrimental effects on a person’s well-being and independence, and high healthcare costs related to treatment and follow-up care.

Normal age-related changes in gait, such as reduced walking speed, decreased hip extension and reduced power generation at the ankle, have been well established in older adults [[3], [4], [5], [6]]. However the gait patterns of individuals with low BMD or osteoporosis have not been thoroughly investigated, and when examined, often studied in relation to falls and kyphosis [[7], [8], [9]] or temporal-spatial gait parameters only [10,11]. In one study, older women with low BMD exhibited increased gait variability in both step and stance time when compared to their peers with healthy BMD levels [10]. Increased gait variability has been associated with greater falls risk in older adults [12], but it has also been reported to improve local dynamic stability when walking slowly in young adults [13]. Temporal-spatial data do not examine the complex interactions between joint movements and mechanical power outputs by the muscles acting about the hip, knee and ankle joints during walking [14]. There is limited evidence of altered gait profiles for older women with low BMD including less hip extension, reduced power generation at the hip in early stance (termed H1 power burst [15]) and pre-swing (H3), less knee power absorption during terminal swing (K4) and smaller ankle power generation at push-off (A2) in comparison with women with normal BMD [16]. Nevertheless, there is a dearth of knowledge on the gait biomechanics of older women with low BMD and osteoporosis. This is important because walking is a recommended form of physical activity for the treatment of low BMD, particularly at the hip and femur [17] as the musculoskeletal loading can promote bone formation and attenuate further bone loss [18,19].

Previous studies investigated how much of the variance in BMD was explained by the peak hip joint moments in all three dimensions during level walking using multiple regression analysis [[19], [20], [21]]. In these studies, 40% to 93% of the femoral BMD variance was explained by hip joint moments [19,20]. However, the results from another study [21], which analysed the hip joint moments independent of body mass, did not fully corroborate with those previous studies [19,20] as the relationship between hip joint moment and BMD was reduced to essentially zero when scaled allometrically. Therefore, this relationship warrants further investigation. This information is important to enable the development of effective, safe and timely exercise-based interventions for older women.

To the authors’ knowledge, there has been only one study to date which has explored the three-dimensional gait kinematics and kinetics in women with osteopenia [16]. However, the participants’ gait speed was not reported [16] and yet the significant effects of speed on gait variables is well established [22]. Therefore, it is unclear how much of the explained variance could be attributed to speed alone in their study [16]. The aim of the current study was to investigate the relationships between gait parameters and self-selected, comfortable walking speed and BMD in older women with a broad range of T-scores (healthy to osteoporotic). It was hypothesised that walking speed would explain most of the variance in gait parameters but that adding T-score into the regression model would contribute to the explained variance for joint kinetic parameters about the hip and knee.