Bone, 11,67-73 (1990) Printed in the USA. All rights reserved.

8756-3282190 $3.00 + .OO Copyright 0 1990 Pergamon Press plc

Sex Differences in Age-related Changes in Vertebral Body Size, Density and Biomechanical Competence in Normal Individuals LIS MOSEKILDE*

and LEIF MOSEKILDE**

* Department of Connective Tissue Biology, Institute of Anatomy, University of Aarhus, and ** Division of Endocrinology,

Department

of Internal Medicine,

Aarhus County Hospital, Aarhus C, Denmark

Address for correspondence

and reprints: Lis Mosekilde, M.D., Department of Connective Tissue Biology, Institute of Anatomy, University of Aarhus, DK-8000 Aarhus C, Denmark.

In the spine, the strength of each vertebral body depends on its size (Biggemann et al. 1988; Brinckmann et al. 1989) the cortical ring and the central trabecular bone mass and structure (Arnold 1966; Atkinson, 1967; Rockoff et al. 1969; Lindahl 1976; Mosekilde and Mosekilde 1986; McBroom et al. 1985). In a previous study (Mosekilde and Mosekilde 1986) we demonstrated that the cortical ring contributes almost the same absolute value throughout life, and that the age-related decline in vertebral body strength mainly depends on the changes in the central trabecular bone mass and structure. Furthermore, it was shown that vertebral bodies increase in size with age due to continuous periosteal growth-and that this could, to some extent, compensate for the unavoidable loss of bone density with age. The investigation was carried out on 44 normal individuals, without separation of males and females. Based on a study of human femora, Martin and Atkinson (1977) concluded that while bone material strength in both sexes declined about equally with age, only men showed changes in other structural properties that could compensate for loss of material strength. Specifically, the second moments of area of femoral diaphyseal cross-sections increased with age in men but not in women. In a study based on human femora and tibiae, Ruff and Hayes showed that while both men and women undergo endosteal resorption of bone and medullary expansion with aging, only men exhibit concurrent subperiosteal bone apposition and expansion (Ruff and Hayes 1988). Thus only men appear to remodel (sic) long bones in a way that would tend to compensate for loss of bone material strength with aging. From this, and from a former study by the same authors (1983), it was deduced “that relatively low activity levels may not stimulate optimal bone remodeling throughout life and thus may contribute to a higher risk of fracture in old age.” As the vertebral body is the main-and first-site for age-related osteoporotic fractures, it would be of major importance to ascertain whether any sex differences in vertebral body size, and changes in this with age, could be demonstrated. The aim of this study was therefore to investigate whether any sex differences in vertebral body size-and in age-related changes in vertebral body size and densitycould be demonstrated in a larger sample (90) of normal

Abstract Size, ash-density and biomechanical competence were investigated on whole vertebral bodies (L,) from 90 normal individuals (47 males and 43 females), aged 15-91 years. At all ages, cross-sectional area was significantly greater in males than in females. Furthermore, in males a significant increase of 25 - 30% in cross-sectional area was demonstrated with aging (r = 0.33, p < 0.05). Conversely, no age-related change in crms-sectional area was detected in females (r = 0.03, n.s.). A significant and identical age-related decrease (p < 0.001) in apparent ash-density was found for both males and females. Biomechanlcal compression tests revealed significant and identical decreases @ < 0.001) in vertebral body load and stress with age in both males and females. However, because of their greater cross-sectional area and an increase in this with age, the level for the load-values was higher in men than in women up to the age of 75 years @ < 0.05). The present study has demonstrated that in men there is a significantly greater cross-sectional area and a significant increase in vertebral body size, due to continuous perlosteal growth. This could, to some extent, compensate for the unavoidable loss of vertebral bone density and stress with age. No age-related compensatory mechanism could be demonstrated in women. Key Words: Vertebral body strength-Bone density-Sex-differences-Aging-Mechanical

size-Bone loading.

Introduction Age-related loss of bone mass, and its clinical consequence of increased fracture risk, has become a major medical problem in western industrialized countries. This has been particularly true for women, although the fracture incidence now seems to increase more rapidly in men than in women-especially for the femoral neck (Johnell et al. 1984; Zetterberg et al. 1984; Liithje 1985). More recently, a similar trend has also been observed for the spine (BengnCr et al. 1988). 67

Li. Mosekilde and Le. Mosekilde: Sex differences in vertebrae

68

individuals. Additionally. the biomechanical consequences of these changes were investigated-with all measurements performed on identical bones (L,). Finally, to separate the contribution of the cortical ring from that of pure trabecular bone, comparison was made between whole vertebral body stress and density and trabecular bone stress and density (L,).

Materials and Methods The study comprised 90 autopsy cases, 43 females aged 15-91 years (mean 62) and 47 males aged 20-90 years (mean 59). The exclusion criteria were diagnoses of malignant diseases and fractures of the removed vertebral bodies (L2 and L,), detected on X rays. None of the individuals was a known alcohol or drug abuser. Some of the elderly individuals had had a period of immobility immediately prior to death (max. 2 months). From most of the individuals, data concerning body heights and weights were available.

Vertebrul trabecular bone (L,) stress and density

To compare whole vertebral body parameters with data from pure trabecular bone, specimens from L, of the same individuals were also investigated. Biomechanical

testing

From the central part of each L3 a vertical trabecular cylinder was obtained (d = 7 mm, h = 5 mm) (Mosekilde et al. 1985). The bone specimens were compressed in a materials testing machine (Alwetron) at a constant compression rate of 2 mm/min. During compression, load deformation curves were obtained, and the stress values were expressed as load divided by cross-sectional area. Measurements

of ash-density

After the testing procedure the trabecular bone cylinders were ashed (2 h at 105°C and 24 h at 58oOC). Ash-density was expressed as ash weight per total sample volume (g/cm3).

Total vertebral body (LJ size, load,

Statistical Analyses

stress and density

Least squares linear regression analysis was used for testing relationships between variables. The significance of differences between mean values was assessed by the Student’s t test. The value p < 0.05 was chosen as the significance level.

From each individual, the second vertebral body (LJ was frozen at - 20°C immediately after removal. The vertebral body was sawed 2 mm below the end-plates to obtain plane-parallel surfaces. By this procedure the nonloadbearing exostoses were removed, and the vertebral bodies comprised only the cortical ring surrounding the central trabecular core (Mosekilde and Mosekilde 1986). Measurements

of cross-sectional

area

The volume of the vertebral bodies was measured by weighing the water displaced when the specimens were suspended in a water bath. The bone marrow was not removed before volume measurement, which therefore determined total volume (bone tissue and marrow). The height of the vertebral bodies was measured with a micrometer, and the mean cross-sectional area (cm*) was expressed as the volume divided by height. Biomechanical

testing

The vertebral bodies, with sawed plano-parallel ends, were compressed in an Instron materials testing machine at a constant rate of 4.5 mm/min. Load-deformation curves were recorded during compression, and from these the maximum load was read in newtons (N) (m x kg x SC*) and the maximum stress, which is load corrected for crosssectional area, was expressed in MegaPascals (MPa) (1 MPa = lN/mm*). Measurements

of vertebral body

apparent ash-density

After the biomechanical testing, the vertebral bodies were ashed in a muffle oven (24 h at 105°C and 24 h at 580°C). The ash-density was expressed as ash weight per unit total sample volume (g/cm3)-and is identical with apparent ash-density.

Results The results are described in 3 sections. First the age-related changes in males and females, then the pure sex-related differences, which are dealt with in 3 discrete age groups, and finally correlation between body height and weight and vertebral body size, load and stress values. I. Age-related

changes

Physical parameters (height and weight). Height: A slight, but not significant decrease

in height was demonstrated for both men and women with age (Fig. 1A). Weight: No difference in weight was seen in women in relation to age. In men, a significant decrease with age was demonstrated (p < 0.01) (Fig. 1B). This decrease was about 15-20 kg from the age of 20 to 80 years. Vertebral body (L,) size, load, stress and density. Vertebral body cross-sectional area: A significant (r = 0.33, p < 0.05) age-related increase in vertebral body

cross-sectional area was demonstrated for men. This increase was in the range of 25-30% from the age of 20 to 80 years. No age-related change in vertebral body cross-sectional area was found for women (r = 0.03, n.s.) (Fig. 2A). Biomechanical competence: The load (strength of the whole vertebral body) declined significantly in both men and women @ < 0.001) (Fig. 2B). The differences between the slopes (- 99.72 * 11.94 versus - 76.94 5 8.05) and the intercepts (11006 2 752 versus 8864 t 535) were not significant. The decline in load values from the age of 20 to 80

Li. Mosekilde and Le. Mosekilde: Sex differences in vertebrae

100

y=-0

127x+181.4

r=-028,

50

n=36

y=-oo95x+l666

50.

NS

r=-030,

n-36

NS

0

10

20

30

40

50

60

70

80

0

90

10

20

30

40

50

60

70

Age,years

30.

30

y-0027x+56

20 t

r=0066

y--0257x+87.1 20.

r=-0.36,

IO-

PC005

n=36

I

0

80 Age

90 years

1 n=36

NS

I 10

20

30

40

50

60

70

80 Age

90 years

0

10

20

30

40

50

60

70

80 Age

90 years

Fig. 1. Relationships between height and age in men and women (A), and between weight and age in men and women (B).

years was 60-65%.

The stress-values

showed

a significant

and identical decrease with age in men and women (p < 0.001) (Fig. 2C). The differences between the slopes (- 0.0783 + 0.0086 versus -0.0640 k 0.0066) and between the intercepts (8.094 k 0.542 versus 7.2% ? 0,435) were not significant. This decrease in stress was 65-70% from the age of 20 to 80 years. Measurements of bone density: The apparent ash-density of the whole vertebral bodies showed a significant and identical decrease with age in men and women. The measured small differences between the slopes (- 0.0013 k 0.0002 versus -0.0011 + 0.0002) and between the intercepts (0.214 * 0.010 versus 0.200 k 0.009) were not significant (Fig. 2D). The decrease in total vertebral body ash density was 35-40% from the age of 20 to 80 years. Vertebral trabecular bone (L,) stress and density. Vertebral trabecular stress: A steady, age-related decrease (p < 0.001) was found for stress values. This decrease was identical for males 0, = -0.0512x + 4.999) and females (y = -0.0553x + 5.292) and was of the magnitude 75-80% from the age of 20 to 80 years. (Fig. ZC-dotted line). Vertebral trabecular ash-density: The ash-density showed exactly the same decrease (p < 0.001) with age in males (y = -0.0016x + 0.222) as in females (y = -0.0016x + 0.216). This decrease was 45-50% from the age of 20 to 80 years. (Fig. 2D-dotted line). 2. Sex-related differences in selected age groups To detect specific sex-related differences related to the menopausal state and advanced age, the females were sub-

divided into three age groups: below 50 years, 50-75 years and over 75 years. These groups were then compared with age-matched males. Below 50 years (Table Z). Body height and weight were sig-

nificantly lower for females than for males. Cross-sectional area of the vertebral body was significantly smaller for females than for males (p < O.Ol), and furthermore, a signiticant difference was found concerning load-values (p < 0.05) between the sexes. No sex differences could be demonstrated concerning stress or density measurements. 50-75 years (Table ZZ). Again, the physical parameters (height and weight) and the cross-sectional area of the vertebral body were significantly smaller for females than for males (p < 0.001). In this age group, too, the load values were significantly lower for females than for males (p < 0.01). No sex-related difference was seen concerning stress or density measurements. Over 75 years (Table ZZZ).The only significant differences in this age-group were a lower height and weight and a smaller cross-sectional area of the vertebral body (p < 0.001) in females. In this group of elderly individuals, no significant differences in load values could be demonstrated between men and women. 3. Correlation between body height and weight and vertebral body size, load and stress For neither males nor females was any significant correlation found between body height or weight and the vertebral body biomechanical parameters (size, load and stress).

Li. Mosekilde and Le. Mosekilde: Sex differences in vertebrae

70

A 18.0

0

9

0

I

8.0

> z

60’

Lj

40.

r” >”

20

y-0.0354x

6.0.

+12.99

r-0.33,

n-46

y=OOO38x+12

19

r-0.052,

4.0

pco.05

n=43

N.S. 2.0

0

10

20

30

40

50

60

70

80

90

Age,

years

0

10

20

30

40

50

60

70

60

90

Age

years

Br x 1000 P 10.

y=-76.94x+6664 r=-0.63,

n-43

pco.oo1 6.

4’

2

I

0

I

0

10

20

30

40

50

60

70

80 Age

90

0

10

20

30

40

50

60

70

80

90

Age

years

years

C.

00

o y=-0.0640x+7 y=-0.0763x+8.094 r=-0.81,

r=-084,

296 n=43

n-46

2.0

10. . .

0

10

20

30

40

50

60

70

60 Age

90

0

10

20

30

40

50

80

70

years

80 Age

. .

‘.

90 years

2 0.10. --._

F t 0.05.

y--00011x+0200

;

P

Sex differences in age-related changes in vertebral body size, density and biomechanical competence in normal individuals.

Size, ash-density and biomechanical competence were investigated on whole vertebral bodies (L2) from 90 normal individuals (47 males and 43 females), ...
782KB Sizes 0 Downloads 0 Views