| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Center for Bone Research at the Sahlgrenska Academy (CBS) (M.L., D.M., C.O.), Department of Internal Medicine, Gothenburg University, 413 45 Gothenburg, Sweden; and Hormone Laboratory (E.H.), Aker University Hospital, N-0514 Oslo, Norway
Address all correspondence and requests for reprints to: Mattias Lorentzon, M.D., Ph.D., Division of Endocrinology, Department of Internal Medicine, Gröna Stråket 8, Sahlgrenska University Hospital, 413 45 Gothenburg, Sweden. E-mail: mattias.lorentzon{at}medic.gu.se.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: The aim of this study was to determine whether smoking was associated with aBMD or volumetric BMD (vBMD) and bone size in young men.
Design and Setting: aBMD was measured using dual x-ray absorptiometry. vBMD and bone size were measured using peripheral quantitative computerized tomography (pQCT). Smoking habits were assessed using questionnaires. Levels of sex steroids, PTH, and 25-OH-vitamin D were measured in serum.
Participants: The population-based Gothenburg Osteoporosis and Obesity Determinants (GOOD) study includes 1068 young men, age 18.9 ± 0.6 yr (mean ± SD).
Main Outcome Measure: The main outcome measure was smoking as predictor of bone parameters and serum sex hormone levels.
Results: Of the study subjects, 8.7% smoked daily. Bone parameters were compared between smokers and nonsmokers. Smokers had significantly lower aBMD (dual x-ray absorptiometry) of the total body (crude: –2.1%; adjusted for age, height, weight, calcium intake, and physical activity: –1.8%), lumbar spine (crude: –4.3%; adjusted: –3.3%), and trochanter (crude: –6.6%; adjusted: –5.0%) than nonsmokers. Using peripheral quantitative computerized tomography, we found that smokers had lower cortical thickness of both the radius (crude: –2.8%; adjusted: –2.9%) and tibia (crude: –4.5%; adjusted: –4.0%) than the nonsmokers, whereas no difference was seen for cortical vBMD. Smokers had higher serum levels of total and free testosterone and lower 25-OH-vitamin D than nonsmokers. Adjustment for testosterone and/or 25-OH-vitamin D levels did not alter the associations between smoking and bone parameters.
Conclusions: We demonstrate that smoking was associated with lower aBMD and reduced cortical thickness in young men.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
111), have shown a negative association between smoking and BMD in young men (10, 11) and women (12), whereas other reports (13, 14, 15) have failed to demonstrate such a relationship. The association between smoking and BMD has been suggested to be gender dependent. The negative impact of smoking during adolescence was in one report found only in men but not in women (10). The degree of smoking has also been implicated to play an important role in the association between smoking and BMD. In a few studies of young men and middle-aged women, only heavy, but not moderate, smoking was associated with reduced BMD (11, 16). In elderly smokers (both men and women), in whom the effect of the cigarette smoke has accumulated over several years, BMD has been found to be reduced, bone resorption increased, and the fracture risk elevated (17, 18, 19, 20, 21).
All previous studies investigating the association between smoking and PBM have been limited by the use of two-dimensional measurements (areal BMD, aBMD) of the bone, using the dual energy x-ray absorptiometry (DXA), a technique that does not provide information about volumetric BMD (vBMD) and size of the cortical and trabecular bone compartments. Thus, due to the limitations of the DXA methodology, it has not been determined whether the presumed effect of smoking on aBMD is due to a decrease in the vBMD or a diminution of bone size of the different bone compartments. In the present large population-based study of 1068 young men [the Gothenburg Osteoporosis and Obesity Determinants (GOOD) study], we have measured the vBMD and the bone size of the cortical and trabecular bone using peripheral quantitative computerized tomography (pQCT), and we investigated the association between smoking and these bone parameters. Furthermore, we have also analyzed potential confounders, including physical activity, calcium intake, levels of sex hormones, vitamin D, and PTH, to determine the independent role of smoking on bone parameters.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The GOOD study was initiated with the aim of determining both environmental and genetic factors involved in the regulation of bone and fat mass. Study subjects were randomly identified using national population registers and were contacted by telephone and invited to participate in the present study (1); 1068 men between the ages 18.0 and 20.1 yr from the greater Gothenburg area were included. To be included in the GOOD study, subjects had to be between 18 and 20 yr of age and willing to participate in the study. There were no other exclusion criteria. A total of 48.6% of the contacted study subject candidates agreed to participate and were included in the present study. A standardized questionnaire was used to collect information about amount of present physical activity (hours/week), physical activity type, nutritional intake (dairy products), and smoking. Subjects who presently smoked regularly (at least one cigarette per day) were categorized as smokers and all other subjects as nonsmokers. Pack years, which equals smoking one pack of 20 cigarettes each day for a year, were calculated from years of smoking and daily consumption. Passive smoking was not investigated. Calcium intake was estimated from dairy product intake. Physical activity type was categorized into peak strain score, as previously described (22). Activities with jumping actions (e.g. basketball) were assigned strain score 3; activities involving repetitive sprinting and turning (e.g. soccer) were given a strain score of 2; all other types of weight-bearing physical activities were assigned a strain score of 1, and nonweight-bearing physical activity (e.g. swimming) a strain score of 0. We have previously analyzed the role of physical activity type and amount in the GOOD cohort (8). The GOOD study was approved by the ethics committee at Gothenburg University. Written and oral informed consent was obtained from all study participants. To determine whether the GOOD cohort was representative of the general young male population in Gothenburg, we compared the height and weight of the GOOD subjects with a group of 624 aged-matched, randomly selected conscripts (86% of all Swedish men undergo testing for military service) living in the same area as the GOOD subjects. There was no difference in height, weight, or body mass index (using an independent samples t test) between these two cohorts, demonstrating that the GOOD cohort is representative of the general young male population of Gothenburg.
Baseline data (of the whole GOOD cohort) concerning bone parameters, anthropometrics, and smoking frequency have been published previously (1).
Anthropometrical measurements
Height and weight were measured using standardized equipment. The coefficients of variation (CV) values were less than 1% for these measurements.
DXA
The aBMD (g/cm2) of the whole body, trochanter, and femoral neck (of the left leg), and lumbar spine were assessed using the Lunar Prodigy DXA (GE Lunar Corp., Madison, WI). The CVs for the aBMD measurements ranged from 0.5 to 3%, depending on application. The CVs for total body lean mass and total body fat mass were 1.8% and 3.4%, respectively.
pQCT
A pQCT device, using single energy x-ray (XCT-2000; Stratec Medizintechnik, GmbH, Pforzheim, Germany) was used to scan the distal leg (tibia) and the distal arm (radius) of the nondominant leg and arm, respectively. The pQCT was calibrated every week using a standard phantom and once every 30 d using a cone phantom provided by the manufacturer. A 2-mm-thick single tomographic slice was scanned with a voxel size of 0.50 mm. The cortical vBMD (mg/cm3), cortical bone mineral content (mg/mm), cortical cross-sectional area (mm2), and cortical thickness (mm) were measured using a scan through the diaphysis (at 25% of the bone length in the proximal direction of the distal end of the bone) of the radius and tibia.
The cortical vBMD is the true cortical volumetric BMD, not including the marrow compartment. Trabecular vBMD (mg/cm3) was measured using a scan through the metaphysis (at 4% of the bone length in the proximal direction of the distal end of the bone) of these bones. All the pQCT analyses were performed by one technician using one pQCT. The CVs were less than 1% for all pQCT measurements.
Serum analyses of sex steroids and SHBG
Analyses of serum levels of total estradiol, testosterone, and SHBG were performed as previously described (23). Free testosterone and free estradiol were calculated according to the method previously described by Vermulen et al. (24) and van den Beld et al. (25). The total estradiol assay CVs were 3% intraassay and 6% interassay. The total testosterone assay CVs were 6% intraassay and 6% interassay. The SHBG assay had an intraassay CV of 3% and an interassay CV of 7%.
Serum analyses of 25-OH-vitamin D and PTH
Serum levels of 25-OH-vitamin D were measured by a competitive RIA (DiaSorin, Stillwater, MN). This assay measures both 25-OH-vitamin D3 and 25-OH-vitamin D2. Intraassay and interassay CVs were 6% and 18%, respectively. Serum levels of intact PTH were measured by an immunoluminometric assay (Diagnostic Products Corp., Los Angeles, CA). Intraassay and interassay CVs were 5% and 9%, respectively.
Statistical analysis
Values are given as mean ± SD unless otherwise indicated. All calculations were performed with the SPSS Statistical Software (version 13.0; SPSS, Chicago, IL). Differences between smokers and nonsmokers were calculated using independent samples t test. The independent predictors of the various bone parameters were tested using multiple linear regression analysis, including height, weight, age, calcium intake, smoking, and physical activity. Smokers were coded as 1 and nonsmokers as 0. Standardized ß-values were used. P < 0.05 was considered significant. Bone parameters (Figs. 1
and 2) were adjusted for calcium intake, amount of physical activity, age, height, and weight using linear regression equations.
|
|
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
There were no differences in age (years), height (centimeters), weight (kilograms), total body lean or fat mass (kilograms), or calcium intake (grams per day) between the nonsmokers and smokers. However, the nonsmokers were significantly more physically active than the smokers (hours per week) (Table 1). Smokers consumed between 1 and 30 cigarettes per day. The mean daily consumption in this group was 9.3 ± 6.3 cigarettes. The mean duration of smoking was 4.1 ± 2.1 yr.
|
). Smoking and aBMD
Smokers had significantly lower aBMD of the total body (–2.1%), lumbar spine (–4.3%), femoral neck (–5.3%), and trochanter (–6.6%) than nonsmokers (Table 1). Because the nonsmokers were more physically active than the smokers, we included physical activity amount, as well as other confounders such as age, height, weight, and calcium intake, in a multiple regression analysis to determine the independent role of smoking for aBMD. In this regression analysis, we found that smoking was an independent predictor of aBMD of the total body, lumbar spine, femoral neck, and trochanter (Table 2). When using adjusted BMD values (for calcium intake, amount of physical activity, age, height, and weight) smokers had significantly lower aBMD of the total body (–1.8%), lumbar spine (–3.3%), femoral neck (–3.9%), and trochanter (–5.0%) than nonsmokers (Fig. 1).
|
The cortical bone size (cortical thickness) of the tibia was smaller in smokers than in nonsmokers (Table 1). The cortical thickness was lower (–4.5%) due to greater endosteal circumference (2.5%) in the smokers than in the nonsmokers, whereas periosteal circumference was not significantly different between the groups (Table 1). For the radius, smokers had lower cortical thickness than the nonsmokers, due to greater endosteal circumference, whereas periosteal circumference was not affected (Table 1). Trabecular vBMD of the tibia but not of the radius was lower in the smokers than in the nonsmokers, whereas no difference was seen for the cortical vBMD of either the tibia or the radius (Table 1).
We used a multiple regression analysis to determine the independent role of smoking for cortical bone size and vBMD of the trabecular and cortical bone compartments. In this regression analysis, we found that smoking was a negative independent predictor of cortical thickness (of both the radius and tibia) due to affected endosteal circumference (Table 2). Smoking was an independent predictor of the trabecular vBMD of the tibia but not of the radius, whereas it did not independently predict cortical vBMD of the radius or the tibia (Table 2). When using adjusted values (for calcium intake, amount of physical activity, age, height, and weight) for cortical bone size parameters of the tibia, smokers had lower cortical thickness (–4.0%) due to greater endosteal circumference (2.7%) than the nonsmokers, whereas periosteal circumference was not significantly different between the two groups (Fig. 2, A–D). Smokers had lower trabecular vBMD (–3.8%) of the tibia, whereas no difference in cortical vBMD could be observed (Fig. 2, E and F).
Smokers had 2.9% lower cortical thickness (2.93 ± 0.26 vs. 2.85 ± 0.26; P < 0.01) and 4.5% greater endosteal circumference (24.7 ± 2.8 vs. 23.7 ± 3.2; P < 0.01) of the radius than the nonsmokers, after adjustment for covariates.
The association between smoking and bone parameters is not influenced by sex steroid levels, 25-OH-vitamin D, or PTH
We investigated whether inclusion of serum levels of free testosterone, free estradiol, 25-OH-vitamin D, or PTH in the regression analysis (together with age, height, weight, calcium intake, and physical activity) influenced the association between smoking and bone parameters. Inclusion of either of these serum parameters separately (data not shown) or all together as covariates did not change any of the associations found between smoking and aBMD (total body ß = –0.06, P = 0.01; lumbar spine ß = –0.07, P = 0.01, femoral neck ß = –0.08 P = 0.006; trochanter ß = –0.08, P = 0.004) or cortical thickness (radius ß = –0.10, P = 0.001; tibia ß = –0.09, P = 0.001).
The association between smoking and bone parameters is not influenced by type or amount of physical activity
As shown, smoking was a negative independent predictor of aBMD and cortical thickness, also after adjustment for present physical activity amount (Table 2). Not only present physical activity amount but also physical activity type categorized according to peak strain score have been shown to be a strong predictor of several bone parameters in this cohort (8). Of the 1068 men, 678 were physically active and 390 were sedentary. The role of smoking on bone parameters was analyzed in the subgroup of 678 physically active men. In this group, 34 men smoked and 644 men were nonsmokers. There was no significant difference in peak strain score between the smoking and nonsmoking men (1.68 ± 0.7 vs. 1.56 ± 0.7; P = 0.36). To investigate whether strain type affected the association between smoking and bone parameters, we next included peak strain score in the regression model (together with the other covariates as presented in Table 2) to evaluate the predictive role of smoking on bone parameters in the group of physically active men. Smoking was still a negative independent predictor of aBMD (lumbar spine, ß = –0.09, P = 0.01; femoral neck, ß = –0.08, P = 0.02; trochanter, ß = –0.11, P < 0.01), and of cortical thickness (radius, ß = –0.11, P < 0.01; tibia, ß = –0.08, P = 0.03).
Smoker subgroup analysis
We analyzed the magnitude and duration of smoking in the group of 93 smokers by converting the amount smoked into pack years (equals smoking 20 cigarettes per day for 1 yr). In the group of smokers, neither years of smoking nor pack years was significantly correlated to aBMD at any site (lumbar spine: years of smoking, ß = –0.14, P = 0.19; pack years, ß = 0.01, P = 0.96, and data not shown) or cortical thickness (tibia: years of smoking, ß=–0.07, P = 0.51; pack years, ß = –0.06, P = 0.61, and data not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Increasing evidence indicate that smoking is associated not only with reduced aBMD in both elderly men and in postmenopausal women but also with fracture risk in these populations (17, 18, 19, 27). Because aBMD in the elderly, the population that sustains the most fractures, is highly dependent on PBM (3, 4), it is of importance to identify risk factors for impaired PBM. The role of smoking in determining PBM has been controversial. Bernaards et al. (14) could not detect that 36-yr-old male and female smokers had lower aBMD of the total body or lumbar spine than the corresponding nonsmokers of the same age. The low number of smokers (n = 36) in this study could have influenced the results. However, Valimaki et al. (10) showed that smoking was an independent negative predictor for aBMD of the lumbar spine and femoral neck in young men but not women (age 20–29 yr). The authors indicate that the observed gender difference could be attributable to that the men smoked a larger number of daily cigarettes than the women. The results from that study should be interpreted with some caution due to the low number of study subjects (111 males and 153 females). In agreement with those findings, recent meta-analyses (including subjects of various ages) showed that smoking was more strongly associated with aBMD and fracture in men than in women (21, 28).
To the best of our knowledge, the present study constitutes the largest population-based study yet performed investigating the relationship between aBMD at the age of PBM and smoking. We demonstrated that smoking was associated with aBMD of the total body, lumbar spine, trochanter, and femoral neck, and that these associations remained after adjustment for body constitution parameters and lifestyle factors, including physical activity amount and type, as well as calcium intake.
Our results showed that smoking was associated with quite substantially lower aBMD, especially at the femur. At this site, the smokers had between 5.3% (femoral neck) and 6.6% (trochanter) lower aBMD than the nonsmokers in our cohort.
Although smokers in our cohort were less physically active, linear regression, analysis including physical activity and other covariates revealed that smoking was a negative independent predictor of aBMD. Interestingly, these findings demonstrate that the possible negative effects of smoking on aBMD can be detected already at a very young age in males. However, even though we adjusted for both physical activity amount and type in our regression analysis, it cannot be totally excluded that the differences in bone mass between smokers and nonsmokers are, to at least some extent, dependent on different physical activity levels between the groups.
Because of the limitation in the DXA methodology, it still remains uncertain whether the association between smoking and aBMD in young adulthood is due to a negative effect on the bone size or on the vBMD. Vehmas et al. (29) recently revealed that smoking was associated with a lower metacarpal cortical index, a measurement of cortical bone mass based on cortical thickness of the metacarpals. With the use of pQCT, we showed in the present study that smoking was a negative independent predictor of the cortical thickness of both the radius and tibia and that this association was due to affected endosteal, but not periosteal, circumference of these bones.
We recently showed that age in the GOOD cohort was associated with an endosteal contraction of the long bones (radius and tibia) (1). Our novel results from the present study suggest that smoking opposes this age-dependent endosteal contraction, resulting in reduced cortical thickness in the smokers.
The mechanism by which smoking affects bone metabolism and bone mass remains inadequately elucidated. Nicotine, the principal pharmacologically active component of cigarette smoke, has been investigated in relation to bone cell function. It has been shown to have direct effects on osteoblast cell proliferation, mediated by specific receptors, and to be able to induce expression of the bone matrix protein osteopontin (30), suggesting direct toxic effects of nicotine on bone cells. In an experimental study using rats, nicotine was shown to cause a reduction in femoral ultimate load and vertebral bone mineral content, whereas no effect of nicotine was seen on tibial cortical or trabecular bone turnover. In our cohort, smoking was not associated with cortical vBMD of either the tibia or the radius, indicating that smoking does not affect the degree of mineralization of the cortical bone in young adulthood.
Several other mechanism, including antiestrogenic effects (31), effects on body weight (28), hypercortisolism (32), and diminished calcium absorption (33), explaining the negative effect of smoking on bone tissue, have been previously suggested. In the present study, body weight was not different between smokers and nonsmokers, and the negative predictive role of smoking on aBMD and cortical thickness was not abolished when adjusting for body weight as a covariate, supporting that the possible effects of smoking on these bone parameters in our cohort were not mediated by body weight. In addition, we also showed that lean mass and fat mass were not different between the smokers and nonsmokers, indicating smoking does not affect body composition in young males. Furthermore, no difference in serum estradiol levels was observed between smokers and nonsmokers, indicating that the associations between smoking and bone mass shown here were not caused by any systemic antiestrogenic effects of cigarette smoke. In agreement with previous findings, the smoking men in our cohort had higher levels of testosterone than the nonsmoking men (34). However, inclusion of testosterone levels in our regression analyses showed that testosterone levels did not influence the association between smoking, aBMD, and cortical thickness, suggesting that the possible effect of smoking on bone parameters is independent of serum testosterone levels. As previously shown in other cohorts (20), smokers in our cohort had lower levels of serum 25-OH-vitamin D levels than nonsmokers. Inclusion of 25-OH-vitamin D levels in our regression analysis indicated that the association between smoking, aBMD, and cortical thickness was independent of vitamin D status.
A limitation of the present study is the cross-sectional design. Furthermore, it should be pointed out that the present study does not completely take all lifestyle factors (that could interact with the association between smoking and bone mass) into account, e.g. alcohol intake was not controlled for.
Previous reports, primarily in elderly populations, have indicated that the association between smoking and aBMD is dose dependent (11, 16, 28). However, subgroup analysis of the smokers in the present study did not reveal any dose-dependent association between smoking and any bone parameter. One may speculate that the relatively low number of subjects in the group of smokers, the small variations in daily cigarette consumption, and short average period of smoking history among the smokers in our cohort could have influenced these results.
The present study is the largest study investigating smoking in relation to aBMD, vBMD, and bone geometry parameters (using both DXA and pQCT) at the age of PBM in males.
In conclusion, we demonstrate that smoking is associated with a reduced aBMD in men at the age of PBM, and that this association is mainly dependent on a reduction in cortical thickness. This knowledge could be of importance when outlining public health recommendations to maximize male bone health in young adulthood.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure Statement: The authors have nothing to disclose. All authors have no conflicts of interest.
First Published Online October 31, 2006
Abbreviations: aBMD, Areal BMD; BMD, bone mineral density; CV, coefficient of variation; DXA, dual energy x-ray absorptiometry; PBM, peak bone mass; pQCT, peripheral quantitative computerized tomography; vBMD, volumetric BMD.
Received June 16, 2006.
Accepted October 23, 2006.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. Sneve, N. Emaus, R. M. Joakimsen, and R. Jorde The association between serum parathyroid hormone and bone mineral density, and the impact of smoking: the Tromso Study Eur. J. Endocrinol., March 1, 2008; 158(3): 401 - 409. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Compston Smoking and the Skeleton J. Clin. Endocrinol. Metab., February 1, 2007; 92(2): 428 - 429. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
