The Effect of Cigarette Smoking on the Development of Osteoporosis and Related Fractures

, Department of Orthopaedic Surgery, University of Pittsburgh, Pittsburgh, Pa.


Medscape General Medicine. 1999;1(3) 

In This Article

Normal Bone Metabolism in Children and Adults

Bone mass in adults, at any age, depends on the peak bone mass attained by the young adult and subsequent bone loss that occurs throughout adult life. During the normal aging process, both bone mass and bone strength decrease, but a variety of genetic and lifestyle factors can influence the rate of bone loss. Smokers are thinner, are less likely to exercise, have poor nutritional habits, and have a reduced absorption of calcium.[7,21,41]

Peak Bone Mass

The bone mass of any individual is determined by the size of the bones and by their mineral density. The bone mass of boys and girls is similar at birth, and only during pubertal maturation do sex differences become apparent, when both the size of bones and the density of bone increase dramatically.[55] There is a four- to six-fold increase in the rate of accretion of bone in the axial skeleton in both males (aged 13-17 years) and females (aged 11-14 years) and a doubling in the appendicular skeleton.[55] Peak bone mass is attained during adolescence and remains relatively stable until the fifth decade of life. The peak bone mass of young men is usually greater than that of young women due primarily to their greater bone size, but BMD is identical in the two sexes after adjustment for body size.[56]

There is wide variation in peak BMD in any population, and current evidence suggests that as much as 80% of this variability may be due to genetic factors.[55] However, it is clear that body weight, nutrition, physical exercise, and endocrine factors also affect bone development in adolescents.

Rate of Bone Loss

Slow bone loss (0.7% to 1% per year) begins in both men and women during the fifth or sixth decade of life[12,57,58,59,60] and occurs equally from both trabecular and cortical sites. During the slow phase, both men and women lose 20% to 30% of both trabecular and cortical bone.[12,59] There is an increased rate of bone loss superimposed on the slow phase of bone loss in women at the start of menopause. This is particularly acute in trabecular bone (spine and calcaneus)[58] and is linked to the characteristic decline in estrogen levels, because ovarian estrogen production slows and then ceases during the perimenopausal and postmenopausal periods. This accelerated phase in women usually lasts for 5 to 10 years[59] and results in a loss of 20% to 30% of trabecular bone and 5% to 10% of cortical bone (Figure 3).

Figure 3. Bone loss in older men and women

Estrogen replacement therapy prevents or slows the bone loss associated with menopause and aging.[61] Similarly, among men who are hypogonadal or who undergo long-term corticosteroid therapy, the administration of testosterone usually increases BMD.[62,63]

The slow phase of bone loss accelerates with increasing age at least up to age 75 to 80 years.[15,64,65] Several recent reports have suggested that this bone loss may be associated with the endogenous levels of sex steroids.[66,67,68,69] Greendale and colleagues[66] noted a strong association between the level of bioavailable estrogen and BMD in both elderly men and women. Moreover, higher bioavailable testosterone (but not total testosterone) levels are associated with higher BMD at the distal radius, spine, and hip among the men and with BMD at the distal radius among the women. More research is needed to clarify the mechanisms involved, but these recent findings stimulated Riggs and colleagues to propose a "unitary model for involutional osteoporosis."[68] Under this unitary model, estrogen deficiency is responsible for both the rapid and slow phases of bone loss in postmenopausal women and for the slow bone loss in older men. Testosterone deficiency also plays a role in the bone loss occurring in men, but the relative importance of estrogen and testosterone levels is not known.

Effects of Cigarette Smoking on Bony Tissues

Cigarette use is believed to have an adverse effect on most aspects of bone metabolism: decreased peak bone mass, increased rate of bone loss, and osteoporosis. To date, there has been only one study that looked at the effect of smoking on peak bone mass in children and young adults.[70] Smoking did not affect BMD in young female twins (average age, 16 years) in this study, but the statistical power to detect an effect was low. Clinical researchers continue to hypothesize that the use of tobacco by teens and preteens will lower the maximal bone mass attained and that the accretion of cortical and trabecular bone may be affected differentially.[12,71]

Epidemiologic Cross-Sectional Studies of BMD

Most epidemiologic studies have found that BMD in older male and postmenopausal female smokers is significantly lower than that in nonsmokers, even after the effects of age, body mass index, and other lifestyle factors have been accounted for.[8,9,72,73,74,75,76,77] The results obtained for BMD at various sites within the body differ across the various studies. In the study by Egger and associates[72] of English men and women aged 61 to 73 years, the lumbar spine BMD was 7% to 8% lower in both male and female smokers compared with the never-smokers. At the femoral neck, the differences were smaller and not statistically significant. However, male and female smokers in the Rancho Bernardo Study (aged 60-99 years) had significantly lower BMD at the hip (but not at the spine or radius) than the nonsmokers.[9] In the Dubbo Osteoporosis Epidemiology Study, tobacco use among men and women (average age, 70 years) was associated with a 5% to 8% reduction in BMD at both the hip and spine.[78]

Results from both the Rancho Bernardo Study[9] and the Dubbo study[78] indicated that cessation of smoking was associated with a BMD level between that of never-smokers and current smokers. These data suggest that cessation of smoking may be helpful in slowing or preventing bone loss even in the elderly.

Smoking seems to have little effect on the BMD of premenopausal women.[13,30,79,80,81] A recent Spanish study of premenopausal women did report an 8% lowering of hip but not spine BMD associated with smoking more than 20 cigarettes per day.[82] However, when the effects of age and body weight were considered, smoking was no longer a determinant of BMD.

An Australian study by Hopper and Seeman[14] of female twins discordant for tobacco use has generated some of the most important information regarding the effects of cigarette use on BMD. Twin research allows precise matching within the twin pairs on age, sex, and genetic composition. With these major determinants of BMD removed from consideration, it becomes possible to examine more precisely the effects of lifestyle factors, such as smoking, on BMD.

Hopper and Seeman measured BMD at the lumbar spine, femoral neck, and femoral shaft in 41 pairs of female twins, aged 27 to 73 years, discordant by at least 5 pack-years of smoking. Among the 20 pairs who were discordant by 20 or more pack-years of smoking, the BMD in the heavier smoking twin was 9% lower at the lumbar spine, 6% at the femoral neck, and 6.5% at the femoral shaft. When all 41 twin pairs were considered, a discordance of 10 pack-years of smoking was associated with a 2% decrease in BMD at the lumbar spine and a 1% decrease at the femoral sites. After controlling for the effects of lifestyle factors (estrogen replacement therapy, oral contraceptive use, exercise, use of alcohol, weight, age at menopause, etc), these differences were not substantively changed.

Hopper and Seeman point out that the clinical significance of these relatively small decrements in BMD may be substantial. For instance, a woman who smokes 1 pack of cigarettes per day during her adult years will have a 5% to 8% lower BMD at the time of menopause than a woman who has never smoked. They estimate that a 10% decrement is equivalent to a 44% increased risk for hip fracture.[14]

Rate of bone loss. Among both healthy postmenopausal women[77] and men of similar age,[60,83] tobacco use is clearly associated with an increased rate of radial bone loss. Sparrow and associates[83] studied bone loss during a 16-year period among 48 pairs of male twins who were veterans of World War II or the Korean War. Those men who smoked lost more bone than those who did not (8.35% vs 5.78%, P = .03). There was a dose-response relationship, and those who smoked more cigarettes per day lost bone at a greater rate (P <.02).[83]

Recently published results from the Rotterdam Study[15] indicated that annual bone loss at the femoral neck was approximately double among the 1856 men and the 2452 women 55 years and older who were current smokers. These results were adjusted for the effects of potential confounders, such as age, body mass index, use of alcohol, dietary calcium, energy intake, and lower-limb disability.

After controlling for years since menopause and body mass index, Dawson-Hughes and colleagues[58] and Krall and Dawson-Hughes[77] reported a 0.9% annual radial bone loss among smokers and no loss among nonsmokers (P <.05) during a 2-year period. They found similar but statistically nonsignificant trends at axial and other appendicular sites, perhaps because their study included only a small number of smokers (n = 34) and therefore had relatively low statistical power to detect differences between the smokers and nonsmokers. An earlier study, also with only small numbers of smokers, failed to show an effect of smoking on bone loss in perimenopausal and postmenopausal women.[84]

Mechanism of Action of Cigarette Smoking on Bone Metabolism

Bone is a dynamic tissue that is continuously remodeled throughout life by the processes of resorption, new formation, and mineralization. Physical forces (exercise, body weight) and hormonal factors (estrogen, testosterone) stimulate osteoclasts to begin the process of bone resorption. Osteoblasts then move into the resorption cavity and form new bone. Mineralization occurs more slowly during the next 3 to 4 months. Cigarette use has been postulated to exert an adverse effect on all three phases of the remodeling (Figure 4), but research to date has not clarified which, if any, of the thousands of breakdown products of burning cigarettes directly affects the osteoclasts and osteoblasts.

Figure 4. Possible mechanisms for effect of smoking on bone mass and fractures

Direct Effects on Osteoclasts and Osteoblasts

A number of in vitro studies using cell cultures have shown that nicotine and free radicals released when cigarettes are smoked act to inhibit the cellular metabolism of osteoblasts and to reduce collagen synthesis. Clinical studies with biomarkers in premenopausal and postmenopausal women[14] have suggested that cigarette smoking affects bone resorption rather than bone formation. These effects appear to be mediated by the action of estrogen on osteoclasts.

Influence of Smoking on Estrogen Metabolism

It is well-established that estrogen plays an important role in the maintenance of bone mass in adult women.[59,61,68,85,86] Smoking appears to have an antiestrogenic effect on the body, reducing the level of circulating estrogens in the body. The rate of breakdown and hydroxylation of estrogens in the liver is increased in smokers.[16] Perhaps in part due to this increased metabolism, smokers undergo menopause at an earlier age.[13] Thus, between menarche and menopause, not only do women smokers experience fewer menstrual cycles, but also during those cycles the estrogen produced is broken down or deactivated more rapidly than in nonsmokers.

In men and in women during the postmenopausal years when the ovarian production of estrogen has essentially ceased, endogenous estrogens are produced in adipose tissue by the aromatization of androstenedione. Since smokers typically have a lower body fat mass than nonsmokers, they would be expected to produce less of these endogenous hormones. It is also postulated that a component of tobacco smoke may inhibit the aromatase system, thereby blocking the aromatization process.[47,87] The net effect is that smokers produce less extraovarian estrogens than do nonsmokers.

Effect on Testosterone and Other Hormones

The effect of cigarette smoking on testosterone levels in men is unclear. The most recent observational studies show that, as in women, it is primarily the level of free serum estrogens rather than free testosterone that predicts BMD in older men.[66,67,88,89] Cigarette smoking inhibits the aromatization of androgens to estrogens in the tissues. These findings most likely explain the increased levels of testosterone and androstenedione found in both male and female smokers.

Smoking also results in an increase in cortisol secretion,[90] hyperthyroidism,[91] and resistance to the action of calcitonin.[92] Although all these effects have the potential to lower BMD in smokers, there is no good evidence that they are physiologically important determinants of BMD.

Effect on Bone Blood Flow

Cigarette use has a number of well-known systemic effects on the cardiovascular system and is a particularly strong risk factor for peripheral vascular disease.[93] Smoking damages the endothelial cells of blood vessels and reduces the production of prostacyclin,[94] thereby reducing blood flow and increasing the risk of thrombosis. To date, there is little experimental evidence linking blood flow and bone metabolism, although it has been postulated that such vascular disturbances may result in low BMD,[95] impaired fracture healing,[96,97,98] and the development of osteonecrosis of the femoral head.[99]