Estrogens and the intervertebral disc

Introduction

In the 18th century, the Scottish surgeon John Hunter discovered that while new bone is laid down in the body, old bone is resorbed. This process, known as remodelling, plays a critical role in osteoporosis, even though it was not recognized as a disease until more than 100 years after his death. In the 1830s, the French pathologist Jean Georges Lobstein noticed that some patients' bones were riddled with larger than normal holes, and he coined the term osteoporosis (porous bone). A century later, Albright ¹ made the observation that postmenopausal women were particularly susceptible to having frail bones. In the 1940s, he began treating women with that condition with estrogen. Twenty years later, Fleisch discovered compounds known as bisphosphonates that inhibit bone resorption. In the meantime, other researchers discovered that selective estrogen receptor modulators could simultaneously block breast tumours and increase bone density.

It was finally in 1984 that the National Institutes of Health published this disease, citing it as a significant threat to health and emphasizing that bone loss could be reduced by estrogen therapy, calcium, good nutrition and exercise. ² Albright ³ had already described osteoporosis as a protein disorder in his original work. However, emphasis on the point gained ground in the 1980s, ⁴ such that the axis of interest slowly started to move from considering bone simply as being solely mineral to considering it as a composite structure of mineral and connective tissue, mainly collagen and glycosaminoglycans. Estrogen plays an important role in the regulation of collagen metabolism ⁵ in postmenopausal women. Connective tissue response to estrogen seems to be universal with positive changes to estrogens being described and the opposite to estrogen deprivation, in different sites such as skin, carotid arteries and bone. ^5–7 Estrogen deprivation leads to a diminution of the collagenous bone matrix. Overall, lack of estrogen is related to increased fracture risk.

Recent work has shown that the intervertebral disc, another site that is predominantly composed of connective tissue, is also similarly negatively affected by the lack of estrogens and positively affected by their presence.

Biomechanics of the intervertebral disc

The intervertebral disc shows progressive changes throughout life. ⁸ Compressive load in the spine is transmitted from vertebra to vertebra via the intervertebral disc. The intervertebral disc is made up of a central nucleus pulposus and an outer annulus fibrosus. These two anatomical elements of each intervertebral disc are made up of varying amounts of collagen, elastin, hydrophilic glycosaminoglycans and water. The intradisc pressure, generated in the fluid-saturated disc tissue enclosed in the thin peripheral fibre layer of the annulus fibrosus, acts to maintain the interspace between the endplates of adjoining vertebrae. Thus, the intervertebral discs act as shock absorbers and are important for maintaining the integrity of the vertebrae. The pressure also tightens the outer fibre layers of the annulus, thus preventing excessive radial bulging.

The intervertebral disc allows for mobility of the spine, and maintains a uniform stress distribution of the area of the vertebral endplates. Also, the disc influences spinal height. The disc tissue is adapted for this biomechanical function. Fibre directions in the annulus fibrosus limit rotational and shearing motion between adjacent vertebrae. The function of the spine is impaired if there is a loss of disc tissue. ⁹ Narrowing of the disc space due to degeneration of intervertebral discs is associated with a significantly increased risk of vertebral fractures. ^10,11

Measurement of intervertebral disc height

In the past, measurement of intervertebral disc height from lateral radiographs could only be performed with limited precision and only for that part of the spine that had been imaged in the central part of the X-ray beam. Normal age-appropriate standard values for disc height were described in a study, ⁹ in which measurement protocols were designed and tested. A database was compiled using measurements from a set of almost 900 lateral views of the lumbar spine of male and female subjects between 16 and 57 years of age. This protocol allows the height and displacement of lumbar vertebrae as well as the height of lumbar discs to be measured for all motion segments on a lateral view, rather than just those vertebrae or discs close to the central beam. This could facilitate recognition of localized abnormalities and retrospective investigations, since the results are independent of exposure geometry. The height of the disc was measured from the ventral height of the disc, measured perpendicular to the bisectrix between the midplanes of adjacent vertebrae. An angle-dependent correction was applied to account for posture-related changes.

According to this study, the relative error in disc height measurement amounted to 4.15%. For example, in the case of a disc of height 10 mm, this corresponds to an error of approximately 0.5 mm. The authors concluded that, on average, the height of lumbar vertebrae is larger in female than in male subjects and the height of lumbar discs is larger in male than in female subjects. In male and female subjects, disc height was noted to be almost constant in the age range between 16 and 57 years. While this was true for female subjects, close inspection of the data indicated that the disc height of male subjects exhibited a very small, but statistically significant, increase with age. For most lumbar discs of male subjects, this height increase amounted to typically 0.0007 units per year. The magnitude of this age-related increase in disc height in male subjects is so small that it cannot be noted by qualitative, visual inspection. The increase, seen in male subjects but not in female subjects, can be explained by the slight age-related rounding of ventral vertebral contours seen in male subjects but again not in female subjects. ⁹ The same group carried out a similar study, this time on the cervical spine instead of the lumbar spine, and concluded that disc height amounts to roughly 1/3 of the vertebral height. Neither men nor women showed a disc height that exhibited age dependence. ¹²

Another study was carried out to examine disc morphology and spinal mobility in subjects with varying degrees of osteoporosis. ¹³ In this case, lateral radiographs and magnetic resonance images of the lumbar spines of 90 elderly subjects with varying bone mineral densities were obtained. The range of motion of the whole lumbar spine was measured using an electromagnetic tracking device. Although the thoracic spine was shown to have decreased anterior vertebral body height in subjects with osteoporosis, this study revealed that the anterior height was increased in the lumbar region. Osteoporosis was associated with expansion of the middle of the disc with corresponding collapse of vertebral bodies.

Nowadays, dual-energy X-ray absorptiometry (DXA) spine scanning is a well-established method used for the measurement of bone density. The spine DXA scan can also be used to assess intervertebral discs by measuring the height of the space between two vertebral bodies. The bone density (g/cm²) of lumbar vertebrae (L2–L4) is normally measured in the supine position, with the legs elevated to minimize lordosis. The measurement of the intervertebral discs can be carried out from the scan image of vertebral bone measurement. On the image, the computer software permits the screen cursors to be moved on to the edges of the vertebrae, measuring the spaces between two adjacent vertebral bodies as the region of interest. ¹⁴

Studies were carried out to evaluate intervertebral disc height by measuring the intervertebral disc space in normal ¹⁵ and postmenopausal women. ^14,16 Gambacciani et al. looked at almost 500 normal women and showed that values of intervertebral disc space were stable from age 20 to 50 years. However, beyond the age of 50 years, there was a significant decrease in the height of intervertebral discs, negatively correlated with both age and years since menopause. In postmenopausal women younger than 60 years, a correlation was evident between intervertebral disc space and years since menopause, but no correlation was evident with age. In women over 60 years, no correlations were found between intervertebral disc space and either age or years since menopause. In three groups of age-matched women, intervertebral disc space was significantly (P < 0.0001) lower in postmenopausal than in both premenopausal and perimenopausal women. ¹⁵

The multivariate analysis done for this study clearly demonstrated that the decrease in intervertebral disc space is strongly related with the menopause. The menopause effect on intervertebral disc space leads to a decrease occuring almost entirely in the first years after menopause. This effect is not linear. Beyond 60 years of age, no further significant decrease in intervertebral disc height was evident. Chronological ageing and other factors seem to play a minor role in the reduction of intervertebral disc space, at least up to the seventh decade in women. This observation suggests that the endocrine and metabolic changes related to menopause, rather than the ageing process, can play a major role in the decrease in intervertebral disc height. This has already been reported for bone mineral density. ^17,18

The effect of hormone replacement therapy (HRT) on intervertebral discs has also been investigated in a number of recent studies. One research group showed that untreated menopausal women had the lowest total disc height, which was significantly lower than the premenopausal group and the hormone-treated group. The second intervertebral disc consistently maintained a significant difference between the untreated menopausal group and the other two groups, the premenopausal group and the treated menopausal group. ^14,16

Thus, women on HRT maintained higher intervertebral discs compared with untreated postmenopausal women. The estrogenic milieu may be relevant because of the significant impact it has on the hydrophilic glycosaminoglycans, the water content, collagen and elastin of the intervertebral discs. The maintenance of adequate disc height may allow intervertebral discs to retain their discoid shape and viscoelastic function, containing vertical forces that may threaten spinal architecture. This could potentially lead to vertebral body compression fractures. Spinal properties, including vertebral macroarchitecture, intervertebral disc integrity, spinal curvature and spinal loading, are thought to be important factors in the ‘vertebral fracture cascade’. ¹⁹

These results are in line with the well-known positive effects of estrogen on collagen metabolism. ^20,21 The intervertebral disc contains the estrogen receptor-b. ²² In addition, 17b-estradiol has been shown to significantly increase the cellular proliferation of cells of the annulus fibrosus when in culture, via the activation of estrogen receptor-b expression. ²² Therefore, estrogen prevents vertebral fractures, not only via its actions on bone but also by maintaining healthy intervertebral discs that can function as shock absorbers, reducing the compression forces on vertebral bodies. Changes in intervertebral collagen could potentially be assessed by measurement of biomarkers, and to what extent such measurements might be clinically useful remains to be determined. ²³

Alternative therapy for the management of osteoporosis in patients in whom hormone replacement is thought to be contraindicated includes bisphosphonates. However, unlike bisphosphonates, estrogen is unique in its ability to affect bone collagen metabolism, apart from suppressing osteoclastic activity. Besides, there is insufficient data on deterioration in bone qualities and the potential of micro-cracks in patients on long-term bisphosphonates. Therefore, estrogen could still be considered as the first-choice therapy for bones and other collagen-rich tissues, because it maintains homeostasis of the bone-remodelling unit. ²⁴ The effects of estrogens on cartilage as demonstrated in the intervertebral disc have not been described with any other antiosteoporosis agent and therefore have to be considered as being of added benefit.

Estrogen has been shown to stimulate osteoblasts in cell culture and increase bone formation in animal models. ²⁵ Estrogen replacement therapy has been shown to have an anabolic effect, which is beneficial to postmenopausal women with osteoporosis. The total collagen content and collagen cross-link maturity were measured in iliac crest bone biopsies in postmenopausal women before and after six years of higher-dose estrogen replacement therapy. The total collagen content of both cortical and cancellous bone increased. Increase in collagen synthesis was supported by a rise in intermediate cross-links in both cortical and cancellous bone and mature cross-links in cortical bone only. ²⁶ At the same time, there was a substantial rise in bone density at both the lumbar spine and the proximal femur. Serum estradiol and bone density results correlated with cortical bone collagen levels. ²⁷ These results suggest that long-term higher-dose estrogen replacement therapy has a therapeutic role due to its anabolic effect on bone in postmenopausal women with osteoporosis.

Conclusion

Menopause has been shown repeatedly to have a negative effect on connective tissue. Such an effect is prevented and in some cases reversed with estrogen therapy. Estrogen has been shown to protect the intervertebral discs as well as other collagen-containing tissues, including bone, carotid artery media and skin. ⁵ Therefore, estrogen replacement is not merely important in the relief of short-term symptoms, particularly in the first decade after menopause, the period when there is the most rapid decline in bone mass and other connective tissue sites. Estrogen has been shown to have profound effects on connective tissue turnover, no matter what the site. This has important implications, not only in maintaining the structure and aesthetic appearance of tissue, but also in maintaining the strength and stiffness of various structures at various sites. The finding that intervertebral discs, predominantly composed as they are of connective tissue, albeit specialized connective tissue, follow the same response to estrogens and the lack of it, as other connective tissues, reinforces the opinion that the estrogen effect is multisystemic and predominantly one of benefit in the systems studied.

Competing interests

None declared.