The main aim was to analyse the associations between several physical fitness variables and bone parameters in a sample of elderly people. 129 participants (94 females and 35 males,
Worldwide life expectancy has more than doubled since 1900, and most people can expect to live into their 60s and beyond [
Aging results in a progressive and generalized impairment of several bodily functions, an increased vulnerability to environmental challenges, and a growing risk of disease and death [
Previous research has studied physical fitness, defined as a set of attributes that are either health- or skill-related [
In relation to elderly people, some of the most widely studied physical fitness-related variables associated with bone health are aerobic fitness, maximal muscle strength, and balance [
Therefore, the aim of this study was to analyse the association between different physical fitness tests and bone variables as measured by pQCT and DXA, in a sample of noninstitutionalized Spanish elderly individuals.
The study was performed in the framework of the elderly EXERNET study (Exernet Elder 3.0), a multicentric study performed between 2008 and 2017 on a representative sample of Spanish seniors from different regions of the country. The inclusion criteria for the EXERNET study were as follows: noninstitutionalized participants over 65 years and not suffering from dementia or cancer, as described elsewhere [
Personal information was collected through a structured and validated questionnaire (which included two specific questions about daily sitting time and daily sedentary hours) [
A portable stadiometer with 2.10 m maximum capacity and 1 mm error margin (Seca 711, Hamburg, Germany) was used to measure height. A body composition analyser with a 200 kg maximum capacity and a ±50 g error margin (TANITA BC-418MA, Tanita Corp., Tokyo, Japan) was used to measure the body mass. Individuals removed shoes and heavy clothes before weighing. Body mass index (BMI) was calculated by dividing weight (kg) by squared height (m2).
Peripheral quantitative computed tomography measurements were taken at four sites (4%, 14%, 38%, and 66%) of the tibia length using a Stratec XCT-2000 L pQCT scanner (Stratec Medizintechnik, Pforzheim, Germany). To ensure machine stability, the pQCT device undertook a daily quality control using a phantom (Stratec Medizintechnik, Pforzheim, Germany).
The nondominant tibia was selected for the measurements. Participants were seated in a stationary chair, adjusted to an appropriate height to ensure the leg was appropriately placed in a straight position. The tibia length from the distal end of the medial malleolus to the medial knee joint cleft was measured. A coronal computed radiograph (scout view) was performed to manually locate a reference line on the distal end of the tibia. The measurement sites were located proximal to this reference line by a distance corresponding to 4% (distal tibia) and 38% (diaphyseal tibia), as previously described [
A DXA scanner (QDR 4500-Explorer, Hologic Corp., Software version 12.4, Bedford, Massachusetts, USA) was used to evaluate aBMD at the lumbar spine (mean
Prior to testing, training workshops were organized to harmonize the assessment of physical fitness among researchers. For this report, six tests modified from the “Senior Fitness Test Battery” (tests: 2, 3, 4, 6) [ Balance test (Flamingo’s test). The maximum standing time (maximum 60 s) on one foot with both hands on the hip was assessed. The test was performed twice with the right and left feet alternatively. The best result obtained among the four attempts was recorded Lower body strength (LBS) test (chair stand test). The number of full stands from a seated position that could be completed in 30 s with arms folded across the chest was determined. This test was performed once Flexibility of the lower extremities (chair sit-and-reach test). The number of centimetres, between the extended and gathered fingers and the tip of the toe (plus or minus, considering if the participants did or did not surpass the tip of the toe, respectively). The test was performed once with each leg, selecting the best attempt for further analyses Agility/dynamic balance (8-foot up-and-go test). Each participant was required to get up from a seated position, walk 2.45 m, and return to a seated position as fast as possible. The test was performed twice and the best result was recorded Maximum walking speed (brisk walking test). This test consisted of a 30 m walking sprint performed as fast as possible. The test was performed twice with at least one minute of rest between repetitions. The best result was recorded Aerobic capacity (6-minute walk test). The distance that participants could walk in 6 minutes around a circuit of 46 m was recorded. Only one attempt was permitted
The Statistical Package for the Social Sciences (SPSS) v. 20.0 for Windows (SPSS, Inc., Chicago, Illinois, USA) was used to analyse the data. All the analyses were performed with the sample divided by sex. Normality of the sampling distribution was assumed as explained by the central limit theorem [
The final sample included 129 participants (35 males and 94 females) aged 65 and older (
Descriptive variables of the sample.
Whole sample ( |
Males ( |
Females ( |
|
---|---|---|---|
Anthropometrics | |||
|
|
|
|
Age (years) |
|
|
|
Height (cm) |
|
|
|
Weight (kg) |
|
|
|
BMI (kg/cm2) |
|
|
|
Fitness variables | |||
Balance (s) |
|
|
|
Lower body strength (reps.) |
|
|
|
Lower body flexibility (cm) |
|
|
|
Agility (s) |
|
|
|
Gait speed (s) |
|
|
|
Aerobic capacity (m) |
|
|
|
pQCT variables | |||
Tt.BMC 4% (g) |
|
|
|
Tt.Ar 4% (mm2) |
|
|
|
Tt.BMD 4% (mg/cm3) |
|
|
|
Tb.BMD 4% (mg/cm3) |
|
|
|
Tt.BMC 38% (g) |
|
|
|
Tt.Ar 38% (mm2) |
|
|
|
Tt.BMD 38% (mg/cm3) |
|
|
|
Ct.BMD 38% (mg/cm3) |
|
|
|
Ct.Th 38% (mm) |
|
|
|
Fracture load X 38% (N) |
|
|
|
SSIp 38% (mm3) |
|
|
|
MCSA 66%(mm2) |
|
|
|
DXA variables | |||
Trochanter aBMD (g/cm2) |
|
|
|
Neck aBMD (g/cm2) |
|
|
|
Ward’s triangle aBMD (g/cm2) |
|
|
|
Total hip aBMD (g/cm2) |
|
|
|
Lumbar spine aBMD (g/cm2) |
|
|
|
Subtotal lean mass (kg) |
|
|
|
SD: standard deviation; reps: repetitions; pQCT: peripheral quantitative computed tomography; DXA: dual-energy X-ray absorptiometry; aBMD: areal bone mineral density; Tt.BMC: total bone mineral content; Tt.Ar: total bone area; Tt.BMD: total bone mineral density; Tb.BMD: trabecular bone mineral density; Ct.BMD: cortical bone mineral density; Ct.Th: cortical thicknes; MCSA: muscle area; SSIp: polar stress strain index.
No associations were found between bone parameters at 4% of the tibia length and physical fitness variables, so they were not presented in the tables.
Regarding 38% of the tibia length, results are presented in Table
Partial correlation coefficients between bone mass variables and physical fitness in males, for age, tibia length, and muscle area as possible confounders.
Balance | LB strength | LB flexibility | Agility | Walking speed | Aerobic capacity | |
---|---|---|---|---|---|---|
pQCT variables | ||||||
Tt.BMD 38% (mg/cm3) | 0.047 |
|
0.202 |
|
|
0.184 |
Ct.BMD 38% (mg/cm3) | -0.141 | 0.295 | 0.198 | -0.346 |
|
0.003 |
Ct Th 38% (mm) | 0.081 |
|
0.218 |
|
-0.175 | 0.261 |
DXA variables | ||||||
Neck aBMD (g/cm2) | -0.184 | -0.313 |
|
0.271 |
|
-0.199 |
Ward’s triangle aBMD (g/cm2) | -0.054 | -0.189 | -0.312 | 0.261 |
|
-0.096 |
LB: lower body; pQCT: peripheral quantitative computed tomography; Tt.BMD: total bone mineral density; Ct.BMD: cortical bone mineral density; Ct.Th: cortical thickness; aBMD: areal bone mineral density. Significant correlations are in bold numbers.
No associations were found between the physical fitness variables and bone tibia variables measured by pQCT for females (all
Neck aBMD was correlated with lower body flexibility (
Balance showed a positive correlation with trochanter aBMD and total hip aBMD (
Partial correlation coefficients between bone mass variables and physical fitness in females, for age, tibia length, and muscle area as possible confounders.
Balance | LB strength | LB flexibility | Agility | Walking speed | Aerobic capacity | ||
---|---|---|---|---|---|---|---|
DXA variables | |||||||
Trochanter aBMD (g/cm2) |
|
0.019 | 0.152 | -0.150 | 0.167 | 0.130 | |
Total hip aBMD (g/cm2) |
|
-0.053 | 0.195 | -0.053 | -0.111 | 0.074 |
LB: lower body; pQCT: peripheral quantitative computed tomography; DXA: dual-energy X-ray absorptiometry; aBMD: areal bone mineral density. Significant correlations are in bold numbers.
Bone variables not showing associations were not shown in tables.
No different results were found when analyses were adjusted by sitting and walking hours for pQCT nor for DXA in neither of the sexes.
Predictive values of fitness in bone variables are presented in Table
Bone mass physical fitness significant predictive values from the stepwise linear regression model for each variable in males and females.
Balance | LB strength | Agility | Walking speed | ||
---|---|---|---|---|---|
Males | |||||
Tt.BMD 38% (mg/cm3) | Overall ( |
— | — | 0.364 | — |
Change |
— | — | 0.248 | — | |
Standardized | — | — | -0.621 | — | |
Unstandardized | — | — | -47.736 | — | |
Ct.BMD 38% (mg/cm3) | Overall ( |
— | — | — | 0.37 |
Change |
— | — | — | 0.195 | |
Standardized | — | — | — | -0.557 | |
Unstandardized | — | — | — | -6.937 | |
Ct.Th 38% (mm) | Overall ( |
— | — | 0.418 | — |
Change |
— | — | 0.230 | — | |
Standardized | — | — | -0.598 | — | |
Unstandardized | — | — | -0.575 | — | |
Females | |||||
Trochanter aBMD (g/cm2) | Overall ( |
0.261 | — | — | — |
Change |
0.042 | — | — | — | |
Standardized | 0.247 | — | — | — | |
Unstandardized | 0.001 | — | — | — | |
Total hip aBMD (g/cm2) | Overall ( |
0.281 | — | — | — |
Change |
0.049 | — | — | — | |
Standardized | 0.265 | — | — | — | |
Unstandardized | 0.001 | — | — | — |
LB: lower body; Tt.BMD: total bone mineral density; Ct.BMD: cortical bone mineral density; Ct.Th: cortical thickness; aBMD: areal bone mineral density. Stepwise regression model controlling for age, object length, and muscle area (for pQCT variables) and age, height, and subtotal lean (for DXA variables) as possible confounders.
Regarding 38% of the tibia length, total bone mineral density and cortical thickness were partially explained by agility (change in
Balance explained trochanter aBMD (change in
No significant different predictive values were found when analyses were adjusted by sitting time and walking hours.
The main findings of the present study were as follows: agility and walking speed showed the greatest influence with bone mass and structure in males, while balance was associated with areal bone mineral density in females.
Due to the possible segmentation of the bone that pQCT provides, different researchers have studied the relationship between bone tissues and several factors like muscle mass [
Bone variables measured by DXA showed a significant association between femoral neck and Ward’s triangle aBMD with agility; however, no significant results were found when applying the linear regressions, which suggest that in males, physical fitness does not influence aBMD variables after accounting for age, height, and lean mass. Results suggest that bone in male elders may have changes in structure not affecting the aBMD [
In females, our data did not present remarkable associations between physical fitness and bone variables measured by pQCT. It is worth considering that most of the published studies examining fitness and bone in elderly population have mainly used DXA, with few studies using pQCT. This limitation makes conclusions derived from previous literature difficult to draw. A previous study using pQCT in females showed that power from the lower limbs predicted a 6.6% of the strength strain index at the tibial mid-shaft [
When evaluating aBMD, analyses did reveal a small predictive contribution of balance to trochanter and total hip aBMD after adjusting by age, height, and lean mass. In this line, controversial results have been found in previous literature. While some studies found a positive correlation between balance and femoral neck [
Flexibility and aerobic capacity did not show associations with bone parameters in our model. A possible reason for this might be that specific physical exercises to improve these variables do not entail high muscle contractions or impacts which could lead to the activation of the osteogenic process. To the best of our knowledge, this study is one of the pioneering studies examining the association between flexibility and bone mass. In relation to the previous research focusing on the relationship between aerobic capacity and bone mass in elderly individuals, unclear conclusions arise from contrasting results. A study with a Portuguese sample of 401 males and 401 females found associations between endurance measured by the 6-minute walk test with hip aBMD in both genders [
The results found in this study show that bone mass is less influenced by physical fitness in females than in males, probably because there are different bone remodelling mechanisms between sexes.
The importance of physical fitness in bone mass found in males when bone was measured by pQCT may be masked in elderly females due to the complexity of female-related issues [
Some strengths and limitations of this study should be highlighted. The present study has a cross-sectional design, reflecting associations but not revealing causality. Further research including larger sample sizes is required to verify these results in representative populations. Although we controlled for several potential confounders, we cannot be certain that other confounders such as dietary calcium intake, smoking, or genetic variations influenced our observations. However, some strengths like harmonized assessments, well-instructed researchers, and validated physical fitness tests should also be considered. Finally, the inclusion of both pQCT and DXA devices to evaluate bone parameters is another strength of this study.
In conclusion, pQCT bone parameters are more influenced by physical fitness in males than females, showing agility and walking the greatest associations. Although DXA is the gold standard diagnosis method for bone health, pQCT should be taken in consideration for a deeper insight of bone and fitness associations in this population.
No data were used to support this study.
The authors declare that they have no conflicts of interest.
The authors are grateful to all the collaborators and volunteers whose cooperation and dedication made this study possible. The elderly EXERNET multicenter study has been supported by the Ministerio de Trabajo y Asuntos Sociales-INMERSO (104/07), the University of Zaragoza (UZ 2008-BIO-01), the Centro Universitario de la Defensa de Zaragoza (UZCUD2016-BIO-01), the Ministerio de Economía, Industria y Competitividad (DEP2016-78309-R), the Biomedical Research Networking Center on Frailty and Healthy Aging of the Centro de Investigación Biomédica en Red Fragilidad y Envejecimiento Saludable, and the FEDER funds from the European Union (CB16/10/00477). A. M. has received a PhD grant from the Gobierno de Aragón (2017-2021).