Greater association of peak neuromuscular performance with cortical bone geometry, bone mass and bone strength than bone density: A study in 417 older women

Highlights

• In 417 older women we evaluated pQCT and DXA measures of bone and neuromuscular performance

• pQCT measures of bone mass, cortical area, cortical thickness and/or bone strength were strongly related to peak neuromuscular performance

• pQCT volumetric bone density did not relate as strongly to peak neuromuscular performance (countermovement jump, grip strength, muscle size)

• This differential was not seen for DXA measures of bone areal mass and density

• The results indicate that, peripherally, neuromuscular performance influence bone geometry and mass more than material bone density.

Abstract

Background

We evaluated which aspects of neuromuscular performance are associated with bone mass, density, strength and geometry.

Methods

417 women aged 60–94 years were examined. Countermovement jump, sit-to-stand test, grip strength, forearm and calf muscle cross-sectional area, areal bone mineral content and density (aBMC and aBMD) at the hip and lumbar spine via dual X-ray absorptiometry, and measures of volumetric vBMC and vBMD, bone geometry and section modulus at 4% and 66% of radius length and 4%, 38% and 66% of tibia length via peripheral quantitative computed tomography were performed. The first principal component of the neuromuscular variables was calculated to generate a summary neuromuscular variable. Percentage of total variance in bone parameters explained by the neuromuscular parameters was calculated. Step-wise regression was also performed.

Results

At all pQCT bone sites (radius, ulna, tibia, fibula), a greater percentage of total variance in measures of bone mass, cortical geometry and/or bone strength was explained by peak neuromuscular performance than for vBMD. Sit-to-stand performance did not relate strongly to bone parameters. No obvious differential in the explanatory power of neuromuscular performance was seen for DXA aBMC versus aBMD. In step-wise regression, bone mass, cortical morphology, and/or strength remained significant in relation to the first principal component of the neuromuscular variables. In no case was vBMD positively related to neuromuscular performance in the final step-wise regression models.

Conclusion

Peak neuromuscular performance has a stronger relationship with leg and forearm bone mass and cortical geometry as well as proximal forearm section modulus than with vBMD.

Introduction

It is well established that bone mineral density (BMD) is related to, or can be modulated by, physical activity [1], [2], local muscular force generation capacity [3], [4], local muscle size [5], [6], [7], [8] and peak explosive power [5]. This has been investigated in a number of populations such as children [5], [6], [8], adolescents [6], young adults [1], athletes [7] and individuals up into their 9th decade of life [2], [3], [4].

However, differences in bone properties are not adequately explained by bone mineral density alone as bone geometry plays an important role [6]. This dependency has been investigated in some cross-sectional studies. For example in pre-pubertal children [5], measures of muscle size and power were related to the derived polar strength-strain index, but not to volumetric BMD (vBMD) as measured via peripheral quantitative computed tomography (pQCT) of the tibial shaft. In female tennis players [7], muscle area was associated with total bone and cortical area and polar second moment of area at the humeral shaft as measured by magnetic resonance imaging and dual X-ray absorptiometry (DXA). In another study of pre-pubertal girls [8], lean tissue mass was the most predictive of bone strength, geometry and areal bone mineral content (BMC) versus other muscle related factors as measured by DXA at the femoral neck. In children, adolescents and adults up to the age of 35 [9], muscle CSA, muscle force, body weight, physical activity were independently associated with polar section modulus, periosteal circumference and cortical area of the tibial shaft but none of the load co-variates were associated with tibial shaft vBMD as measured by pQCT. Whilst the relationship of bone density to physical capacity has been relatively well investigated, further work is needed to better understand the relationship to broader bone characteristics. Specifically, we wanted to better understand which aspects of bone density, mass and geometry are most related to, or most influenced by, neuromuscular function.

In the current investigation we considered the relationship between physical capacity (grip strength, jumping), forearm and calf muscle size to bone mineral density, bone mass, bone geometry and derived measures of bone strength in women aged 60 to 95 years of age. This investigation was conducted on the basis of data collected as part of the larger “Bus-tour” study [10].