Ph.D. thesis. OSTEOPOROSIS IN MEN Odense Feb 2011

1
Ph.D. thesis.
OSTEOPOROSIS IN MEN
Odense Feb 2011
Morten Frost
Endocrine Research Unit, Department of Endocrinology, Odense University Hospital
Institute of Clinical Research, Faculty of Health Sciences, University of Southern Denmark
2
TABLE OF CONTENTS
1. ACKNOWLEDGEMENTS .......................................................................................................................................... 3
2. FINANCIAL DISCLOSURES ..................................................................................................................................... 4
3. SUPERVISORS ............................................................................................................................................................. 4
4. LIST OF PAPERS ......................................................................................................................................................... 5
5. ABBREVATIONS......................................................................................................................................................... 6
6. ENGLISH SUMMARY ................................................................................................................................................ 7
7. DANSK RESUMÉ ........................................................................................................................................................ 9
8. INTRODUCTION ..................................................................................................................................................... 11
9. AIMS ............................................................................................................................................................................ 12
10. DEFINITION OF OSTEOPOROSIS IN MEN .................................................................................................... 13
11. EPIDEMIOLOGY OF OSTEOPOROSIS IN MEN ............................................................................................. 14
12. EPIDEMIOLOGY OF FRACTURES IN MEN .................................................................................................... 15
13. PATHOGENESIS ..................................................................................................................................................... 20
14. PREVENTION ......................................................................................................................................................... 39
15. CONSEQUENCES OF FRACTURES ................................................................................................................... 42
16. PHARMACOLOGICAL THERAPHY.................................................................................................................... 43
17. CONCLUSION ......................................................................................................................................................... 48
18. PERSPECTIVES ....................................................................................................................................................... 50
19. APPENDICES ........................................................................................................................................................... 51
20. REFERENCES ......................................................................................................................................................... 58
21 VITAMIN D STATUS AND PTH IN YOUNG MEN: A CROSS-SECTIONAL STUDY ON ASSOCIATIONS
WITH BONE MINERAL DENSITY, BODY COMPOSITION AND GLUCOSE METABOLISM .................... 76
22 OSTEOPOROSIS AND VERTEBRAL FRACTURES IN MEN AGED 60-74 YEARS ..................................... 98
23 RISK FACTORS FOR FRACTURE IN ELDERLY MEN. A POPULATION-BASED PROSPECTIVE
STUDY........................................................................................................................................................................... 117
24 PATTERN OF USE OF DXA SCANS IN MEN. A CROSS-SECTIONAL, POPULATION-BASED STUDY
........................................................................................................................................................................................ 141
3
1. ACKNOWLEDGEMENT
Above all, I would like to express my deepest appreciation to the staff at the osteoporosis clinic.
Your company and talents are second to none, and it has been a tremendous pleasure to work
together with you all.
My fellow students are sincerely thanked for their excellent company and the many enjoyable times
at home and abroad. The people who designed and conducted the male cohort studies AGSY and
OAS are greatly thanked for their collaboration and willingness to share their data.
I would like to express my profound gratitude to my supervisors - Marianne, Bo and Kim. Thankyou for your kind support and the introduction to the intriguing world of endocrine research. It has
been an enormous pleasure to learn from your immense knowledge of endocrinology and in
particular to discuss politics, food, coffee, German movies and English trains and many, many other
things. For that I’m most grateful.
Someone once told me that supervision was no different from managing a forest: you drop a few
tree seeds and then contemplate the growth-potential of those that germinate. Regrettably, there was
no indication of when the trees would be axed or whether the fate of the wood would be cardboard
or something slightly more notable.
In that framework, my main supervisor - Kim Brixen - also manages woodland, but his generous
and constant support allows a variety of trees to grow without any restrictions (or fear of the axe).
Your enthusiastic pursuit of challenges rather than dreary reflections on difficulties or incapacities,
as well as excellent problem-solving skills and a genuine confidence in other people are impressive
and inspiring. Our discussions have been very rewarding and for my part they have promoted
reasoning and evidence as well as discouraged narrow-mindedness. I highly respect your
knowledge and judgment and I am most grateful for your support as supervisor during my PhD. I
look forward to future conversations and collaboration.
4
2. FINANCIAL DISCLOSURES
This PhD study received financial support from the Velux Foundation and the University of Southern
Denmark. The projects were supported by the Novo Nordisk Foundation, the World Anti-Doping
Agency, the Danish Ministry of Culture and Pfizer Denmark. I am investigator in clinical studies
sponsored by MSD, Amgen and Novartis Healthcare. Finally, I hold stocks in Novo Nordisk.
3. SUPERVISORS
Marianne Andersen, Associated Professor, PhD
Department of Endocrinology, Odense University Hospital
Bo Abrahamsen, Professor, PhD
Department of Internal Medicine F, Gentofte Hospital
Kim Brixen, Professor, PhD (main supervisor)
Department of Endocrinology, Odense University Hospital
Who all have affiliation with:
Institute of Clinical Research, Faculty of Health Sciences, University of Southern
Denmark
5
4. LIST OF PAPERS
Paper 1.
Vitamin D status and PTH in young men: a cross-sectional study on associations with
bone mineral density, body composition and glucose metabolism
Clin Endocrinol (Oxf). 2010 (ONLINE)
Paper 2.
Cross-sectional, population-based study on the prevalence of osteoporosis and vertebral
fractures in elderly Danish men
Accepted for publication, OI
Paper 3.
Risk factors for fracture in elderly men: a population-based prospective study
Accepted for publication, OI
Paper 4.
Pattern of use of DXA scans in men: a cross-sectional, population-based study
Manuscript submitted
6
5. ABBREVATIONS
ADT:
AED:
BFAT:
BMC:
BMD:
BMDhip:
BMDneck:
BMDspine:
BMDwb:
BMI:
BTM:
COPD:
CVD:
DB RCT:
DXA:
E:
ESR1:
FEA:
GC:
GIO:
GWAS:
HR:
HR-pQCT:
LBM:
PBM:
pQCT:
RA:
RR:
SERM:
SHBG:
SNP:
SOST:
SSRI:
T:
TCA:
T2DM:
TCA:
VFA:
VFx:
WHO:
Anti-androgen treatment
Anti-epileptic drug
Total body fat
Bone mineral content
Bone mineral density
BMD total hip (by DXA)
BMD femoral neck (by DXA)
BMD lumbar spine (by DXA)
BMD whole body (by DXA)
Body mass index
Bone turnover markers
Chronic obstructive pulmonary disease
Cardiovascular disease
Double-blinded, randomized, placebo-controlled study
Dual energy x-ray absorptiometry
Oestradiol
Oestrogen receptor alpha-gene
Finite element analysis
Glucocortioid steroid
Glucocorticoid induces osteoporosis.
Genome wide association study
Hazard ratio
High resolution peripheral quantitative computed tomography
Lean body mass.
Peak bone mass
peripheral quantitative computed tomography
Rheumatoid arthritis
Relative risk
Selective oestrogen receptor modulator
Sexual hormone binding globulin
Single nucleotide polymorphism
Sclerostin
Serotonin reuptake inhibitors
Testosterone
Tricyclic anti-depressants
Type 2 diabetes
Tricyclic antidepressants
Vertebral fracture assessment
Vertebral fracture
World Health Organization
7
6. ENGLISH SUMMARY
Fractures are a leading cause of morbidity and mortality in the elderly. One of the major reasons for
fractures is osteoporosis. It has been estimated that in Denmark, 41% of women and 18% of men aged
over 50 years have osteoporosis. While the majority of fractures in men are observed in the young,
fractures are also prevalent in the elderly, affecting approximately 20% of men aged over 50 years in
their remaining lifetime.
This thesis reviews the current knowledge about osteoporosis in men. Moreover, it
incorporates four studies performed in three population-based cohorts of young and elderly men that
focus on aspects of male osteoporosis.
Decreased levels of vitamin D are known to cause rickets in children and osteomalacia
later in life as well as increase fracture risk in the elderly. The prevalence of vitamin D deficiency and
the impact of vitamin D deficiency on peak bone mass, bone markers and other metabolic functions
were evaluated in a cross-sectional study comprising 800 Danish men aged 20-29 years (Paper I). In
summer and winter, 6% and 44% young of the participants were vitamin D deficient, respectively.
Moreover, low levels of vitamin D were associated with significantly lower bone mass in young men at
the age where peak bone mass is attained.
Data on the prevalence of osteoporosis are essential for priority setting and development
of prevention strategies. In Paper II, the prevalence of osteoporosis was evaluated in a populationbased, cross-sectional study of 600 Danish men aged 60-74 years. On the basis of Danish reference
material, the prevalence of osteoporosis was found to be 10%. Additionally, 6% of the men had a
vertebral fracture. While osteoporosis and fractures are closely related, only one in four participants
with a fracture had in fact osteoporosis. Although prevention of these fractures is of key importance,
knowledge about risk factors for fracture in elderly men is limited. In Paper III, predictors of fractures
in elderly men were evaluated in a population-based study of 4,975 Danish men aged 60-74 years
8
followed for 5 years. We found a 68% increased risk of fracture in men with a family history of hip
fractures and a graded relation between number of falls and risk of fracture, particularly among the
elderly and non-obese.
Assessment of bone mineral density and fracture risk is central to fracture prevention and
treatment of osteoporosis. In Paper IV, the pattern of use of DXA and 10-year absolute fracture risk
was assessed in the 4,975 participants included in study III. DXA had been performed in 3% of the
study population. In all, 21% and 10% of those reporting previous DXA and no previous DXA,
respectively, had a 10-year risk of a major osteoporotic fracture above 20%. Conversely, 32% of the
patients previously evaluated by DXA appeared to have no clinical risk factors.
A number of other studies have made a significant contribution to knowledge about
osteoporosis in men. The pathogenesis of the disease is multifaceted; genetics, lifestyle factors,
medication and other diseases and conditions are known to confer an increased risk of osteoporosis
and fractures. Osteoporosis in men differs in a number of ways from that observed in women. First,
fracture incidence is higher in men aged less than 50 years. In contrast, the risk of a second fracture is
similar in men and women. Second, the criterion for the diagnosis of osteoporosis in men remains
debatable. Third, fractures are associated with a significantly higher mortality in men. Fourth,
osteoporosis is less commonly diagnosed and treated in men than in women. Finally, the number of
clinical trials evaluating the effect of osteoporosis-specific treatment on fracture risk in men is very low.
Although our knowledge about male osteoporosis is growing, a number of issues still
remain unanswered, including the reasons for underdiagnosis, inadequate treatment and increased
mortality of men with osteoporosis. Future studies will hopefully provide this information.
9
7. DANSK RESUMÉ
Knoglebrud medfører betydelig morbiditet og mortalitet hos ældre mennesker. Knogleskørhed er en
vigtig årsag til knoglebrud, og i Danmark anslås det, at 41 % af alle kvinder og 18 % af alle mænd over
50 år har sygdommen. Knoglebrud optræder hos mænd hyppigst i ungdomsårene, men 20 % af alle
mænd over 50 år antages at pådrage sig et knoglebrud i deres resterende levetid.
I de efterfølgende afsnit gennemgås den aktuelle viden om knogleskørhed hos mænd.
Desuden inddrages 4 populationsbaserede undersøgelser, som belyser forskellige aspekter af
knogleskørhed hos mænd.
Lave niveauer af D-vitamin kan medføre knoglesygdom. Forekomsten af mangel på
vitamin D hos yngre danske mænd og effekten af en mangel på det maksimale niveau af knoglemassen
belyses i den første artikel, som baserer sig på en undersøgelse blandt 800 danske mænd i alderen 20-29
år. Undersøgelsen viste, at niveauerne af D vitamin var utilstrækkelige hos 6 % om sommeren og 44 %
om vinteren. Desuden blev det påvist, at vitamin D niveauer, desto lavere var indholdet af mineral i
knoglerne.
Viden om prævalensen af en given sygdom er afgørende for både prioritering og
planlægning af forebyggende tiltag. I den anden artikel, som baserer sig på en populationsbaseret
undersøgelse omfattende 600 danske mænd i alderen 60-74 år, fandtes 10 % at have knogleskørhed,
mens 6 % havde sammenfald i ryggen. Denne undersøgelse viste også, det ikke har nogen væsentlig
betydning for estimatet af prævalensen, om man vælger et dansk eller amerikansk normalmateriale i
forbindelse med beregning af forekomsten af knogleskørhed
Den væsentligste kliniske følge af knogleskørhed er knoglebrud. Vores viden om
årsagerne til knoglebrud hos mænd er desværre begrænset. I den tredje artikel undersøges faktorer, der
knytter sig til en øget risiko for knoglebrud. Undersøgelsen baserede sig på en populationsbaseret
gruppe omfattende 4,975 danske mænd i alderen 60-74 år, som blev fulgt i 5 år. Undersøgelsen viste, at
10
risikoen for et knoglebrud var forøget med 68%, hvis der i familien var personer, der havde brækket
hoften. Desuden var der er sammenhæng mellem antal fald og risikoen for knoglebrud, særligt hos
ældre og normalvægtige.
Vurdering af BMD og risikoen for knoglebrud er af stor betydning i forbindelse med
forebyggelse af knoglebrud og behandling af osteoporose hos mænd. I den fjerde artikel blev
anvendelsen af DXA og 10 års risikoen for knoglebrud vurderet i de 4975 mænd, der deltog i det tredje
studie. DXA var udført hos 3 % af deltagerne. I alt 21 % af dem, der havde været undersøgt med DXA,
og 10 % af dem, der ikke havde været undersøgt, havde en 10 års risiko for knoglebrud på mindst 20
%. Omvendt havde 32 % af dem, der tidligere havde været undersøgt med DXA, ingen umiddelbare
kliniske risiko faktorer for knoglebrud.
Flere andre undersøgelser har bidraget med en betydelig mængde information om
mandlig osteoporose. Patogenesen er multifacetteret; både genetiske, livsstilsmæssige, farmakologiske
og konkurrerende sygdomme kan øge risikoen for osteoporose og knoglebrud. Osteoporose hos mænd
adskiller sig på en række punkter fra sygdommen hos kvinder. For det første er forekomsten af
knoglebrud højere hos mænd inden 50 års alderen, mens hyppigheden af knoglebrud efter 50-års
alderen er højest hos kvinder. I modsætning til dette er risikoen for et efterfølgende knoglebrud ens hos
mænd og kvinder. For det andet er de diagnostiske kriterier for osteoporose hos mænd fortsat
omdiskuterede. For det tredje, knoglebrud er fulgt af en betydelig højere dødelighed hos mænd. For det
fjerde, osteoporose diagnosticeres sjældnere og behandles i mindre grad hos mænd. Endelig er der kun
få kliniske studier, der belyser effekten af specifik medicinsk behandling af knogleskørhed hos mænd på
forekomsten af knoglebrud.
Selvom omfanget af viden om mandlig osteoporose er stigende, er en række spørgsmål
uafklarede. Dette gælder fx årsagen til den forøgede dødelighed samt utilstrækkelige diagnosticering og
behandling. Fremtidige studier vil forhåbentlig bidrage med denne vigtige information.
11
8. INTRODUCTION
Life expectancy has increased significantly in the last centuries, and the populations of the developed
countries are aging considerably (1). Fortunately, the processes of aging appear to be changeable, and
disability is not necessarily a hallmark of aging (2). In elderly men, osteoporosis and osteoporotic
fractures are a significant cause of disability and mortality (3). Hip fractures are more prevalent in the
elderly, but the incidence appears to be declining in Danish men (4).
Bone health is not a concern specific to the elderly. At all ages, bone tissue is continuously degraded
and rebuilt in a process known as bone remodelling. Any factor promoting bone resorption without
concomitant bone formation or decreased formation with unchanged resorption reduces bone strength
causing an increase in fracture risk. Issues potentially detrimental to bone health should therefore be
addressed throughout life if osteoporosis and osteoporotic fractures are to be prevented.
Osteoporosis is a disease characterized by low bone mineral density (BMD) and disrupted bone
microarchitecture that impairs the properties of bone and subsequently leads to an increase in the risk
of fracture (5). The World Health Organization (WHO) has operationally defined osteoporosis in men
and women as a BMD of less than or equal to 2.5 standard deviations below the mean of a young
reference (6). Based on register data, 41% and 18% of Danish women and men aged more than 50
years are thought to have osteoporosis (7). The lifetime risk of fractures in women above 50 years of
age is close to 50% while that of men is around 20% (8); the risk of a second fracture is independent of
gender, however (9). In addition, hip fractures in men are associated with twice the mortality of hip
fractures in women (10). The worldwide distribution of fractures is uneven as the prevalence of
osteoporotic fractures including hip fractures is higher in Northern European and Scandinavian
countries (3;11;12). Unfortunately, although treatment options are available, limited awareness on the
part of both patients and clinicians has caused osteoporosis to be underdiagnosed and inadequately
treated in men (13-16).
12
A number of factors are currently known to increase the risk of osteoporosis and osteoporotic fractures
in men. These include both modifiable and permanent risk factors, i.e. glucocorticoids (17), falls (18),
smoking (19) and genetics (20). The relative importance of several of these risk factors has been
substantiated by meta-analyses including almost 15,000 men (21), and the results are now utilized in the
WHO fracture risk algorithm (FRAX©). Contrary to preceding fracture algorithms, FRAX allows for
calculation of absolute 10-year major osteoporotic fracture risk in men. Subsequently, the algorithm has
been validated in approximately 230,000 persons, although only a few hundred of those were men (21).
FRAX may prove useful for identification of high-risk patients, but trials conducted on individuals
selected for treatment on the basis of their 10-year fracture risk are not available.
Since 39% of all fractures occur in men, knowledge about the causes, prevention and treatment of male
osteoporosis should be addressed (3). Current knowledge about osteoporosis and fractures in men is
reviewed in the following sections. This includes results from the four studies on male bone health,
including prevalence of osteoporosis among older Danish men.
9. AIMS
The aims of the present thesis were to
1) Investigate the prevalence of vitamin D deficiency among young Danish men and the impact of
this on bone turnover markers (BTM) and BMD
2) Evaluate the prevalence of osteoporosis in elderly Danish men
3) Assess risk factors for fractures in elderly Danish men
4) Determine the pattern of use of bone mineral density assessment in elderly Danish men
5) To provide a review of osteoporosis in men.
13
10. DEFINITION OF OSTEOPOROSIS IN MEN
The WHO guidelines for the diagnosis of osteoporosis recommend the use of BMD assessed by e.g.
dual energy X-ray absorptiometry (DXA) (6), but osteoporosis is also diagnosed in the event of a
fragility fracture irrespective of BMD (22). A fragility fracture reflects progression of underlying
abnormalities that cause bone loss and acts as a measure of bone status as well as a strong predictor of
fracture. BMD and fracture risks are thus inversely associated in both men and women (23-26). Some
studies suggest that fracture risk is similar in women and men at the same level of BMD (23;25;27).
However, the US Study on Osteoporotic Fractures in Men (US MrOS) and Study of Osteoporotic Fractures (SOF)
reported a larger increase in hip fracture risk for every decrease of one standard deviation of BMD in
men compared to women (Age-adjusted RH (95%CI): 3.2 (2.4-4.1) vs. 2.1 (1.8-2.4)) (24). In the same
studies, fracture risks were lowest in men, whether or not sex-specific or female reference values were
used (24). These differences may be related to increased risk of falls in women or differences in the agerelated deterioration of bone structure between men and women. Nevertheless, application of sexspecific references may be prudent, and they are currently recommended for T-score calculations in
men by the International Society for Clinical Densitometry and the US National Osteoporosis
Foundation (28;29).
The T-score cut-off used for the diagnosis of osteoporosis was primarily selected on the basis of the
observation that the number of individuals with a T-score of less than -2.5 was equivalent to the
proportion of fractures observed in postmenopausal women (6). Currently, the same criterion is
recommended in both sexes.
The validity of osteoporosis in men being defined as a T-score of -2.5 has been questioned. In the
Rotterdam study, the sensitivity of DXA to identify persons who would experience a non-vertebral
fracture was estimated to be 44% and 21% in women and men, respectively (27). While BMD is
predictive of fractures in women and men, focus on risk factors and estimation of the absolute risk of
fractures may prove even more prudent in men.
14
In summary, osteoporosis is currently defined on the basis of measurement of BMD using a cut-off Tscore of -2.5 or less in both sexes. Some, however, also consider a fragility fracture as a criterion for the
diagnosis. Men and women with the same T-score have different fracture risks, whereas the risk of a
fracture is at least as high in men as in women at similar absolute levels of BMD.
11. EPIDEMIOLOGY OF OSTEOPOROSIS IN MEN
The prevalence of osteoporosis depends on the diagnostic criteria as well as ethnic group, age and sex
of the population at interest. Thus, in the National Health and Nutrition Survey (NHANES III), the
prevalence of osteoporosis was estimated to be 3% in Hispanic, 5% in black and 7% in white men (30).
The prevalence of osteoporosis is not uniform, and in particular the Northern European countries are
regarded as high risk countries for osteoporosis (31). In a Dutch study, the prevalence of osteoporosis
was 20% and 40% in men aged more than 70 and 80 years, respectively (27) whereas the prevalence of
osteoporosis was 22% in a Belgian study conducted among men at a mean age 69 years (32). Estimates
are generally not concordant, and even within Scandinavia the reports on the prevalence of
osteoporosis disagree. In a Swedish study, 17% of men aged 80-84 years were considered to have
osteoporosis (33) whereas 18% of all Danish men aged more than 50 years are thought have
osteoporosis on the basis of register-based data and Danish reference BMD (7). Recruitment of study
participants and the mean age of the study populations explain at least in part this variation.
In a population-based, cross-sectional study comprising 600 men aged 60-74 years using DXA and
either Danish or NHANES III reference values for the calculation of T-scores, approximately 10% of
the study population was found to have osteoporosis (Paper 2). Although the number of individuals
considered osteoporotic by use of either of these reference values was similar, there were significant
differences in the prevalence of osteoporosis between each site investigated, e.g. using NHANESIII
only 0.5% had osteoporosis at the total hip whereas the Danish reference resulted in a prevalence of
15
4.4%. The distributions of BMD in the total hip, femoral neck and lumbar spine measured in young
Danish men aged 20-29 years and these older men are presented in Figure 1.
These results show that the prevalence of osteoporosis is highly dependent on the reference values
used and the anatomical region that is investigated. The estimate of osteoporosis in Danish men may
1.6
1.4
.8
1
1.2
BMD: Lumbar spine (g/cm2)
1.2
.4
.6
.6
.8
1
BMD: Fermoral neck (g/cm2)
1.2
1
.8
.6
BMD: Hip (g/cm2)
1.4
1.4
1.6
prove relevant to priority setting.
20
40
60
80
20
Age (years)
40
60
80
20
Age (years)
Figure 1. The distribution of BMD in the total hip, femoral neck and lumbar spine measured in Danish
men aged 20-29 years and 60-74 years. (Paper 2)
12. EPIDEMIOLOGY OF FRACTURES IN MEN
Fracture incidence differs significantly with age in both sexes. In children and until the age of
approximately 50 years, men are more likely than women to sustain a fracture (34). In a British cohort
study comprising 6% of the total population, fracture incidences were higher in boys than girls at all
times of childhood with peaks at ages 11 and 14 years in girls and boys, respectively (35). Men aged 1559 years experienced a 2.9 times higher fracture incidence compared to women in a Scottish study on
15,000 fractures (36), whereas the incidence of all fractures including the occurrence of diaphyseal
fractures of the femur, tibia and forearm was higher in women after the age of 50 years (34;36). These
differences are likely to be caused by several factors related to physical activities, i.e. sports or trauma
(37), however, intrinsic factors also influence risk of fracture. In an English study including more than
40
Age (
16
6,000 children aged 10 years, Clark et al. (38) found the risk of fracture to be inversely associated with
volumetric BMD irrespective of fracture mechanism, suggesting that an underlying skeletal fragility
contribute to the risk of fracture even in children sustaining traumatic fractures. Prospective studies on
the impact of childhood fractures on fracture incidence in adulthood are limited, but at least selfreported childhood fractures have been shown not to increase the risk of fractures in adulthood (39).
The world-wide osteoporotic fracture incidence in 2000 was estimated to be 9 million, of
which 39% reportedly occurred in men (3). The distribution of fractures was uneven as 42% of
vertebral, 30% of hip, 25% of humeral and 20% of forearm fractures were observed in men (3).
Compared to women, men have the lowest fracture incidence at all ages after the age of 50 years
(3;9;12;27;34;36;40;41), and the lifetime risk of fractures in men older than 50 years has been estimated
to be 20% compared to 50% in women (12;41;42). In the Dubbo osteoporosis study (DOES) that has
16 years of follow-up, fracture risks were 1% and 3.2% in men and 3.2% and 5.0% in women aged 6069 and 80+ years, respectively (9). There was, however, no difference in the risk of a second fracture
(men/women: 60-69 years: 5.7% versus 6.2%; 80+ years: 9.0% versus 8.9 %). A greater relative risk of
incident fracture after a low-trauma fracture was reported in men (HR: men: 3.47 (2.68-4.48); women:
1.95 (1.70.2.25)). Due to lower risk of an initial fracture in men, however, absolute fracture risk was
similar in both sexes (9).
In the population-based Tromsø study including 28,000 participants (12), the 10-year absolute risk for
all non-vertebral fractures was higher in men until the age of 45 years. The absolute risk for nonvertebral and osteoporotic fractures exceeded 10% in men aged 65 and 70 years and in women at ages
45 and 50 years. With lifetime risks of non-vertebral and osteoporotic fractures at the age of 50 years
reaching 38% and 25% in men and 67% and 55% in women, respectively, incidence rates reported in
the Tromsø study are among the highest published (12).
17
12a. OSTEOPOROTIC FRACTURES
Generally, fractures are considered osteoporotic if spontaneous or caused by low-energy trauma. Highenergy traumatic fractures, however, were also associated with both reduced BMD and increased risk of
subsequent fractures in the SOF and MrOS studies, and the associations were of the same size as those
between low-trauma fractures and BMD (43). More recently, the SOF and the DOES linked certain
fracture types with osteoporosis (44-46). In women, the risk of a second fracture was the same
irrespective of the classification of the first fracture, and a similar but insignificant trend was observed
in men (43). The number of high-trauma fractures was significantly higher in men than women (21%
vs. 9%) and these were due to sporting and other leisure activities as well as traffic accidents. Hightrauma fractures in women were mainly due to traffic accidents (43). The consequences of these
findings are that osteoporosis should be considered in any case of a fracture, independent of the trauma
mechanism, and that this may be of particular relevance in men.
In a meta-analysis including approximately 60,000 persons (of whom a quarter were men) from
Europe, Australia, Japan, US and Canada followed for a duration of 250,000 person-years, a history of
previous fracture significantly increased the risk of a further fracture (RR: 1.86 (1.75-1.98)) without any
significant sex-related difference in risk ratios (47). The risk of a succeeding fracture remains higher
than in the general population in the first year (48), but studies have generally found accumulation of
fractures after 1-5 years (49-52) and even 10-15 years (9;53). Any fracture observed should therefore be
considered a sign of increased risk of a second fracture and the future fracture risk should be assessed.
12b. TYPES OF FRACTURES IN ELDERLY MEN
Several different types of fractures have been linked to osteoporosis in men. van Staa et al. (42)
reported that the most commonly observed fractures in men were those of the femur, vertebrae,
forearm, humerus, clavicle, scapula, ribs and pelvis (42). Several other research studies or public
18
databases have provided information on fractures in men, including from Scottish (54) and Icelandic
(55) sources.
12c. HIP FRACTURES
The age-adjusted women:men hip fracture ratio varies internationally and is reported to vary between 34:1 in whites to 1:1 in South African blacks (56). Generally, the incidence increases later in men than
women (34). The highest incidence is found among Scandinavian men and white North America men
while the lowest is found among Asians and blacks (31;56-58). In Denmark, the incidence in men and
women aged over 50 years is 2.5/1000 and 6.6/1000, respectively (59).
12d. SPINE FRACTURES
Only 25-33% of vertebral fractures are diagnosed clinically (60;61), but even sub-clinical fractures are
associated with significant morbidity (3) and mortality (62) in both men and women. The prevalence of
thoracic vertebral fractures was evaluated in 30,000 men and 27,000 Finnish women and showed
increased prevalence in men after 40 years of age and in women after 55 years. The prevalence was
higher in men at all ages except after the age of 75 years (63), whereas in the European Vertebral
Osteoporosis Study (EVOS) comprising more than 15,000 participants including well over 7,000 men
aged 50-79 years, the prevalence of vertebral fractures was only higher in men aged under 65 years. The
vertebral fracture incidence was up to three times higher in Scandinavians compared to other
Europeans (64). The European Prospective Osteoporosis Study (EPOS), which consists of almost
3,000 men and women aged 50+ years, found an incidence in men of about half that observed in
women (5.7/1000 vs. 10.7/1000). In the Framingham study, the prevalence of vertebral fractures was
comparable between men and women, but the female vertebral fracture incidence was twice as high as
that observed in men (24% vs.10%) (65).
19
The occurrence of vertebral fractures has been evaluated in the almost 6,000 male participants of the
MrOS. The incidence rate was 2.2/1000 with higher rates in men aged 80+ years. In all, 13% of the
incident vertebral fracture cases were observed in participants with osteoporosis compared to 2% in the
study population, and approximately three-quarters of the vertebral fractures were caused by unknown
factors or low-energy trauma (18).
In the previously mentioned Danish study on men aged 60-74 years, the prevalence of vertebral
fractures as identified by use of vertebral fracture assessment was 6.3% (Paper 2). In accordance with
the findings of the US MrOS, the prevalence of osteoporosis among these fracture patients was low.
These results suggest that DXA is inadequate in identifying those with vertebral fractures and new
methods of identifying vertebral fractures, e.g. VFA, are needed. This necessity is underlined by results
from the UK general practice database showing that the presence of vertebral fractures was associated
with increased mortality in men (42). Similar results have been found in a number of other studies from
Europe, Canada and Australia (66-68). Therefore, identification of vertebral fractures is as important in
men as in women.
12e. FOREARM FRACTURES
Contrary to women, the incidence of forearm fractures does not increase in men until late in life
(27;34;36;69;70). In the Rotterdam study (27), forearm fractures increased in women from the age of 55
years whereas an increase was noted in men only after the age of 75 years. As with other fractures, the
incidence in forearm fractures differs between countries, with the highest occurrence in Norwegians
(70). In men as opposed to women, Colles fractures appear to increase the absolute risk of a
subsequent hip fracture more than a vertebral fracture (71), and the importance of non-hip, nonvertebral fractures is further underlined by the association with a doubling of mortality risk in men after
any of these fractures (68).
20
12f. HUMERUS, RIB AND PELVIC FRACTURES
The incidence of humerus fractures increases modestly with age in men and women (27;34;72),
although substantially more in women than in men (36). Recent data from the MrOS have highlighted
the importance of evaluation of patients with rib fractures (73). Being the most common incident
fracture in participants of the MrOS with an incidence rate of 3.5/1000 patient years, the rib fracture
was, like other fragility fractures, predictive of fractures of the hip, spine and forearm independently of
co-morbid conditions (73).
Data on pelvic fractures in men are limited. Singer et al. (36) found similar incidences in men and
women aged 15-49 years with increases in the subsequent years, particularly in women. Low-energy
pelvic fractures are associated with increased mortality in both sexes, although the increase was higher
in men compared to women in a recent German study of 1,100 patients with pelvic fractures and
almost 6,000 controls (74).
The pattern of fractures observed in men and women differs. Fractures are common in elderly men as
in women, but in men the incidence increases at a later age than in women. Men have a lower incidence
of fractures; however, fractures are associated with increased morbidity and mortality in men as well as
women.
13. PATHOGENESIS
Osteoporosis and osteoporotic fractures are caused by both intrinsic and extrinsic factors. Many of
these may impact both men and women, but the relative importance of the factors differs.
13a. GENETICS
Several studies on twins and family members have assessed the impact of genetics on aspects of bone
biology, including BMD, peak bone mass (PBM), bone loss, bone turnover markers, skeletal size and
21
geometry as well as fractures. Studies suggest that the importance of heritability is largest in the young.
Hereditary factors account for 60-80% of the variation in PBM, irrespective of sex (75;76).
Sex- and site-specific quantitative trait loci that regulate BMD have been found, but the results were
not confirmed in meta-analyses including over 11,000 individuals (77;78). Interestingly, the skeletal size
of the father appears to be more strongly associated with skeletal size in female than male offspring,
further underlining the importance of genetics and that sex-specific effects may be present (79).
Studies on candidate genes have been performed in several cohorts, most of which have included
substantially more women than men. Only few sex-specific effects of common variations in genes on
bone are known. Variations in the oestrogen receptor alpha-gene (ESR1) gene were associated with
vertebral fractures in women (80), whereas a specific variation in the transforming growth factor beta 1
gene appeared to associate with lumbar spine BMD in men only (81). Subsequent meta-analyses on
studies evaluating candidate genes have provided more power to detect significant effects. These
studies have shown that variations in the low-density lipoprotein-related protein 5 (Lrp5) and 6, key
regulators of bone mass (82), the ESR1, which plays a role in bone acquisition/maintenance (80), and
the collagen type 1 alpha 1 gene (83) are associated with BMD and fracture risk.
Genome wide association studies (GWAS) have provided further information on the effect of common
differences in genes on BMD. In the Framingham Study, Kiel et al. (84) showed associations between
several common genetic variations and BMD as well as bone geometry and a number of other
measures of bone status in 1,141 persons including sex-specific associations. Larger and statistically
more powerful multi-national GWAS on BMD and fracture risk were reported from an Icelandic
cohort (n=5,861, 13% men) with replication in Icelandic (n=4,165, 26% men), Danish (n=2,269, no
men) and Australian cohorts (n=1,461, 39% men) (85) as well as a UK study with replication in three
Western European cohorts (n=8,557, 9% men)(20). However, the single nucleotide polymorphisms
SNPs found to be associated with BMD and fractures were not in complete concordance. Nine SNPs
were associated with BMD in a meta-analysis on five GWAS, whereas four were associated with
22
fracture risk, including Lrp5 and SOST, a gene encoding sclerostin which inhibits bone formation.
Neither of the GWAS on fracture reported sex-stratified analyses, and the effectiveness of these
variants to predict fractures remains inadequate for clinical purposes. Although several genes have been
shown to be associated with BMD and fractures, only 4-5% of the variation in BMD is accounted for
by known alleles (86). Possibly, the information provided by these GWAS could be explored if geneenvironment interaction was evaluated. In a Danish study, common variations in the Lrp5 gene were
found to be associated with PBM in non-sedentary men only, suggesting that the effect of the genetic
variation was modulated by physical activity (87).
Despite current uncertainties, the variants discovered in these studies provide knowledge about
biochemical pathways that is useful for the understanding of the pathogenesis of osteoporosis and,
perhaps new targets for treatment of the disease.
13b. FAMILY HISTORY
In the Rancho Bernardo Study (88), a family history of a fracture was significantly related to BMD of
the hip in men and the lumbar spine in women. The authors also found an inverse relation between
BMD and the number of family members known to have osteoporosis. Consistently, a parental history
of fracture was associated with lower BMD at the spine and proximal femur in the US MrOS (89). On
the basis of almost 35,000 persons followed for well over 130,000 person-years, however, Kanis et al.
(90) showed an association between a parental hip fracture and an increased risk of an osteoporotic and
hip fracture in women only (osteoporotic: women vs. men, RR (95%CI): 1.38 (1.16-1.65) vs. 1.01 (0.671.52). Hip: 1.75 (1.17-2.63) vs. 1.73 (0.82-3.63)), although the estimates are approximately of the same
size (91).
In order to evaluate the extent and causes of fractures in elderly Danish men, including the effect of a
family history of osteoporosis or hip fracture on fracture risk, we conducted a questionnaire-based
survey on aspects of health, and in particular osteoporosis, among a random sample of over 9,000 men
23
aged 60-74 years (Study of osteoporosis and male aging (SOMA)). Responders were asked not only to submit
information about lifestyle factors, comorbidities, medication and disposition for osteoporosis and hip
fractures but also to consent to a register-based follow-up of incident fractures. In all, 4,696 returned a
completed questionnaire and these individuals were subsequently included in the study. Responders
were compared to non-responders as well as the complete Danish age- and sex-matched population.
These analyses indicated that those included in the study were healthier, had a higher income and were
more likely to be married. During a follow-up time of 5.4 years, 203 men experienced a first fracture, of
which 85 were considered osteoporotic. Further analyses showed that a family history of hip fracture
was associated with a significantly increased risk of any – but not osteoporotic – fracture (1.68 (1.142.49)). This suggests that heritability or shared environment has an impact on fracture risk in elderly
Danish men as well.
The risk conferred by a parental family history of a fracture, either osteoporotic or hip fracture, appears
to be reduced, if not non-existing, in the oldest old. Parental history of a hip fracture was significantly
associated with an increased risk of an osteoporotic fracture and a hip fracture at ages 60-75 years in
men and women (Osteoporotic fractures: RR age 60: 1.56 (1.22-1.98); RR age 75: 1.31 (1.07-1.61). Hip
fractures: RR age 60: 2.41 (1.03-5.64): RR age 75: 1.75 (1.08-2.82)), whereas in participants aged 80+
years, a parental history of a hip fracture had no effect on the risk of an osteoporotic or hip fracture
(92;93). These results suggest that fracture heritability may diminish with increased age, something that
is likely to be due to greater influence of environment rather than genetics as age increases. These
findings favour the idea that genetics primarily affect PBM and are supported by reports that genetic
effects on bone loss are inconsistent (92;93).
In accordance with these results, a Swedish twin study on fracture risk found an association between
genetics and fracture risk (94). The heritability of a hip fracture was age-related with heritability
estimates in persons aged <70 years of 0.68, in persons aged 70-79 years 0.47 and in those older than
24
80 years 0.03. None of these estimates appeared to be sex-specific. The age of the study population was
50+ years and the heritability of fractures in younger persons remains unknown.
Genetics is important to bone health, but data suggest that hereditary factors are important for fracture
risk in the young and elderly but not the oldest old. Current knowledge about genes that confer disease
is limited, but future studies evaluating not only SNPs but also copy number variations, epigenetics and
other aspects may provide substantiation.
13c. EARLY LIFE
According to the foetal origins hypothesis, birth weight is inversely associated with the risk of
unfavourable health effects in later life, including type 2 diabetes and coronary heart disease (95). The
biological mechanism that may be operating here remains unknown, although common genes may
influence both foetal growth and health in adult life or the foetus may be programmed in response to
maternal malnutrition (96). A number of studies have suggested that there is a similar association
between intrauterine life and adverse bone outcomes in adulthood. Thus, birth weight appears to be
associated with PBM (97;98) and bone mass in adulthood (99). Data from our group, however, suggest
that the association between birth weight and PBM is driven by the association between birth weight
and muscle mass, and then muscle mass and PBM (Frederiksen L et al. Abstract, ECTS 2006). In a
Finnish study including approximately 7,000 individuals persons born in the period 1923-33, low
growth in early life was associated with increased risk of hip fracture (100). In addition, the level of
vitamin D in early life has been associated with bone mineral content at 9 years of age (101). Javaid et
al. (101) suggested that intrauterine lack of vitamin D may affect bone mass by influencing calcium
transportation across the placenta, which may impair intrauterine and postnatal bone growth (101).
Conversely, a recent study on more than 7,000 children showed no effect of maternal pre-pregnancy
body mass index on the child’s bone mineral content, suggesting that other factors may explain the
25
association (102). Further studies are needed for substantiation of the effect of vitamin D on birth
weight as well as the relation between birth weight and bone health in later life.
13d. PEAK BONE MASS
PBM is the amount of bone tissue present at the end of skeletal maturation (103) and is recognized as a
major determinant of bone health in later life (104). The timing of PBM has been debated, and by far
the majority of studies investigating this issue have been performed in women. It is generally thought,
however, that maximum bone mineral content (BMC) and areal BMD (aBMD) in the femoral neck and
lumbar spine are achieved later in men (105). In an 8-year follow-up study on BMD in men aged 17-26
years, Nordström et al. (106) found loss of BMD in the hip from the age of 19 years but stable levels of
total body and lumbar spine BMD during the last part of the study period. In a Danish cross-sectional
study from our group on approximately 800 men aged 20-29 years, PBM had been attained (107).
Although useful as a reference, local data on the level of PBM cannot be employed in other
geographical regions (108).
Studies relying on aBMD may be by weakened by the technique, since aBMD includes information on
bone width and height but not depth. Increases in bone size could be perceived as an increase in
aBMD, even though the true density or volumetric BMD (vBMD) may be stable. Using vBMD rather
than aBMD, Henry et al. (109) showed that PBM is reached in the radius and femoral neck by late
adolescence, whereas vBMD of the spine peaked at 22 years in men and 29 years in women. Compared
to girls, the trabecular number and thickness increased later in boys and while cortical thickness and
density remained stable in boys from pre- to midpuberty, a small decrease was found in girls eventually
causing the finite element models to show reduced bone strength in girls compared to boys (110).
Evaluating the distal radius, young men aged 20-29 years were found to have larger bones and
endocortical areas as well as trabecular thickness and trabecular bone volume/tissue volume ratio,
whereas there was an equal number of trabeculae, trabecular separation and cortical thickness (111).
26
The absolute level of bone mass and bone strength is higher in boys at the time of PBM. By the age of
65 years, PBM may still account for half the variation in BMD (112), and the importance of PBM is
emphasized further by a doubling of the fracture risk for every single decline of one standard deviation
of BMD (113). Although genetics explain most of the variation in PBM (75;114), several factors known
to influence PBM are potentially modifiable by preventive strategies such as physical activity (115),
protein and calcium intake (116), and vitamin D supplementation (117). Factors favouring gain of bone
mass in early adulthood could, therefore, be of great significance to bone health in later life.
13e. AGE-RELATED CHANGES IN BONE MASS AND MICROARCHITECTURE
BMD is a significant determinant of fracture risk in men, and assessment of bone mass after
achievement of PBM is used clinically (24;118). Several studies have consistently associated advancing
age with lower BMD of at the hip, albeit lower in men compared to women (119-122). The grade of
reduction in bone mass differed in various age groups in the MrOS. Although the mean loss of BMD
was 1.72% during 4.6 years of study participation, those older than 85 years experienced a 2.5 times
higher reduction in the femoral neck BMD compared to those aged 65 (123). The effect of age on
spinal BMD has been debated. In the Framingham Study (119), BMD of the spine was reduced
substantially less than observed in women, whereas no change was found in NEMO (Network in
Europe for Male Osteoporosis), a European study consisting of 1300 men aged 50-86 years (124). A
UK study reported increases of spinal BMD (121) in men, whereas lumbar spine BMD in Japanese men
increased in the younger decades and diminished in the oldest (125). Other age-related factors probably
influence the evaluation of spinal BMD as indicated in the French minos study where significant loss of
spinal bone mass was evident only after exclusion of cases of arthritis (122).
vBMD of the spine, femoral neck, distal radius and distal tibia decrease from early adulthood in both
sexes, but the magnitude is larger in women (126). With age, the trabecular thickness decreases more in
men than women, whereas trabecular number lessens and the trabecular separation increases in women
27
(111). In a cross-sectional study of men and women aged 20-99 years, total bone area as well as
trabecular number and thickness were higher in men and the age-related reductions were comparable in
both sexes, with the exception of distal radius trabecular thickness that appeared to be reduced to a
larger extent in men (127). In addition, cortical porosity was more extensive in young men than women,
although porosity increased substantially more in women than men (127).
Cortical thickness and vBMD in the distal tibia and femoral neck in men and women remain stable
until the age of 50 years. Afterwards, substantial decreases – albeit significantly greater at the femoral
neck in women – of these two parameters are observed (111;128). Accordingly, femoral neck vBMD
was 22% lower in men aged >85 years than in men aged 65-69 years as observed in the MrOS study.
The femoral shaft cross-sectional area increased by 9%, and the cortical area was reduced by almost 6%
due to substantial periosteal apposition and increase of endocortical area (129).
Various tests of biomechanical properties and the novel finite-element analysis (FEA) allow for
computer-based estimation of changes in bone strength and have been used to assess sex- and agerelated changes. Distal radial and tibial bones are stronger in men as assessed by FEA, and distal radius
carries significantly more load in women than men (130). Women experience the largest reduction in
vBMD of the femoral neck, but the loss of femoral bone strength is even greater and significantly
higher than that observed in men (131). The ratio of fall load and bone strength in the femoral neck
and distal forearm appears to worsen more in women than men. Vertebral strength is higher in men
and remains so with age, primarily due to a larger cross-sectional area (132). Cortical porosity is more
strongly correlated with age than any other measures of bone architecture obtained by HR-pQCT, and
changes in cortical porosity confer considerable impairment of the biomechanical properties (133).
Collectively, the studies assessing bone density, architecture and strength by use of DXA, pQCT, HRpQCT and FEA have contributed significant information about the potential reason for the observed
differences in fracture incidences. Thus, the age-related reductions in femoral neck, vertebrae and distal
radius bone strength could, at least in part, explain the substantially lower fracture risk observed in men
28
since loss of trabeculae seen in women causes larger decreases in bone strength than thinning of
trabeculae in FEAs (134). Further studies are needed to examine the usefulness of pQCT, HR-pQCT
and FEA for fracture prediction.
13f. SEX STEROIDS
Sex steroids are essential for growth and preservation of bone in men and women (135). Total, free and
bioavailable sex steroids decrease and sex hormone-binding globulin (SHBG) increases with age in men
(136;137).
In men, anti-androgen treatment (ADT) has been used to study the effect of testosterone (T) and
oestradiol (E) on bone. In young men on ADT, T was found to regulate bone formation while
resorption was affected by both T and E. In contrast, in elderly men E influenced bone resorption
while both E and T affected bone formation (138;139). In the Rancho Bernardo study (136), higher
bioavailable T and E in particular were associated with BMD in men, while Slemenda et al. (140) found
E to be positively and T negatively associated with BMD in a 2-year prospective study on men with an
average age of 65 years. The European Male Ageing Study (141), comprising 3141 men aged 50-79
years, showed significant association between bone quality assessed by ultrasound and free and total E
but not T. In the MINOS study (142), low serum levels of E were associated with increased bone
turnover markers and lover BMD, whereas in the Framingham study (143;144), hypogonadism
appeared not to influence bone health or hip fracture risk unless matched with low E; E was, however,
strongly associated with bone mass and increased hip fracture risk. In the MrOS, T and E deficiency
were found in 6.9% and 9.2% of osteoporotic participants, respectively, and rapid bone loss was most
often found in men with T deficiency (145). Nevertheless, low E, high SHBG or the combination of
low bioavailable T and high SHBG were associated with an increased risk of osteoporotic fractures in
the MrOS (146).
29
Contrary to these results, the DOES (147) showed that T and not E was associated with fracture risk
and the Tromsö study (148) showed that neither T nor E was linked to fracture risk in men. In elderly
men, the threshold for increased bone loss due to E has been estimated to be approximately 114
pmol/l (149). Local thresholds may have to be developed, however, as study populations and methods
used for measuring sex steroids are likely to influence the results.
Hypogonadism due to ADT causes considerable bone loss at cortical and trabecular sites (150). In a
study on 50,000 men surviving at least 5 years after being diagnosed with prostate cancer, 19.4% of
those who received ADT and 12.6% of controls experienced a fracture. Risk of fracture at the spine,
femoral neck, rib, humerus, forearm and pelvis was increased in the ADT group (151). Corresponding
increases in fracture risk were shown in a Danish register-based case-control study of prostate cancer
patients (152).
Sex steroids are key elements in the pathogenesis of male as in female osteoporosis. The increased
fracture risk during ADT underlines the importance of T, but E appears to have the strongest
association with BMD as well as fracture risk in population-based studies.
13g. BODY WEIGHT AND WEIGHT LOSS
In adults, body weight and body mass index (BMI) have been directly associated with BMD. On the
basis of a number of previously published studies, Papaioannou et al. (153) suggested that for every
extra 10 kg of body weight, BMD of the hip and spine increased by 3-7%. Conversely, low body
weight, BMI and weight loss are predictors of bone loss at the hip and spine of approximately the same
magnitude in both sexes (154). The increase in fracture risk conferred by low BMI has been shown to
depend on BMD but not age and sex (155). In a Norwegian study, loss of forearm BMD during 7 years
of follow-up was significantly higher in slim participants compared to those with normal weight (156),
and in the MrOS increasing body weight was associated with higher total hip BMD of 0.1%/year while
those undergoing intentional weight reduction faced an annual bone loss of 1.4% (157). Several large
30
studies have confirmed the association between BMI and fracture risk (119;158-160). The relation
between BMI and BMD is not linear, nor is it fully explained. Thus, after adjustments for BMD and
falls, obesity was associated with an increased risk of fracture in the MrOS, which included 72%
overweight or obese participants (159). In a Swedish study covering >30,000 middle-aged men and
women, high BMI was associated with increase risk of humerus and ankle fractures and lower risk of
forearm fractures (161). In the SOMA study, self-reported weight loss was associated with a higher risk
of any fracture after adjustment for age, although weight loss was not significantly associated with
fracture risk after adjustments for other risk factors (HR (95%CI): 1.35 (0.88-2.06)).
The mechanisms explaining the association are largely unknown. The effect of increased fracture risk
may be caused by lower levels of vitamin D in obese individuals as well as reduced physical activity and
more falls (162;163). While mechanical stress and the effect of load on bone may mediate a substantial
part of the observed association between relative weight and bone mass, a recent study on analyses of
femoral strength suggested that lean body mass and not fat-mass accounted for the association (164).
BMI appears to influence bone status equally in men and women. Low BMI is an established risk factor
for fractures, but due to the increasing prevalence of obesity (2), fracture prevention may have to target
both under- and overweight individuals.
13h. PHYSICAL ACTIVITY
Physical activity is generally considered beneficial to bones. In young men, bone appears to adapt to
changes in levels of physical activity (165), and while weight-bearing activities promote bone gain (166),
sedentary lifestyle impairs bone status due to detrimental effects on particularly trabecular bone (167).
Although the beneficial effects diminish after cessation of physical activity, a net gain of bone mass
appears to remain at least in young men and women (168). Elderly men and women may also benefit
from physical activity. In a 10-year prospective study, continuing physical activity as opposed to a
sedentary lifestyle maintained bone mass in elderly men and women (169). In a Norwegian study,
31
leisure physical activity was associated with significantly higher levels of BMD at both weight and nonweight bearing skeletal sites after an observation time of 22 years irrespective of sex (170). In addition,
using the “Up and Go” test, Cauley et al. (89) found a 2-4% higher BMD in those capable of getting up
whereas Hannan et al. (119) showed a doubling of the bone loss in patients spending most of the day in
bed.
Paganini-Hill et al. (171) showed in a study population of >8,000 women and >5,000 men that at least
one hour compared to less than half an hour of exercise per day was associated with significantly
reduced risk of hip fracture in both sexes. A similar result on hip fracture risk was observed in a
Finnish study of 3,000 men followed for 21 years (172). Conversely, while physical activity was
associated with a reduced risk of vertebral fracture in the EVOS study covering in excess of 5,000 men
and 6,000 women, men belonging to the top quartile of physical activity had the highest risk (173),
indicating that exercise may not uniformly improve bone status.
There may be multiple causes for the observed association between physical activity and bone status.
Being physical active may promote balance causing fewer falls, improve muscle strength or stimulate
bone formation directly. Although data on potential fracture prevention is limited in men, physical
activity – and especially weight-bearing activity – appears to improve bone status. Nonetheless, future
studies evaluating the effect of exercise on fracture risk in men are needed. Current knowledge about
the type and amount of exercise, effect size, duration of effect and potential adverse effects is
incomplete, and randomized studies evaluating the effect of exercise on fracture risk are needed.
13i. VITAMIN D
Vitamin D acts primarily by increasing calcium absorption in the intestine and bone mineralization.
Secondary hyperparathyroidism may develop in cases of vitamin D inadequacy, increasing bone
turnover and subsequently bone loss and fracture risk (174). In the young, only few studies have
evaluated vitamin D status and the relationship between vitamin D status and PBM in men. In a French
32
study, approximately 70% of boys aged 13-16 years had vitamin D levels below 25 nmol/l during
winter (175), whereas the prevalence of vitamin D deficiency (defined as 25OHD <20nmol/l) in 220
Finnish men (primarily army recruits) aged between 18.3 and 20.6 years was 40% during winter (176).
In the latter, reduced levels of BMD in the lumbar spine and total hip were reported in vitamin Ddeficient participants (176).
In a Danish population-based sample of 800 men aged 20-29 years, the prevalence of vitamin D
deficiency defined as serum 25-OHD below 50 nmol/l, as suggested in a recent report (177), was 44%
during winter and 6% in the summer (Paper II). Participants with a sufficient level of vitamin D had
significantly higher levels of BMD at the femoral neck, total hip and spine compared to the remaining
study population, suggesting that vitamin D status may be improved in otherwise healthy young men.
BMI, visceral fat mass, subcutaneous fat mass and smoking were inversely associated with vitamin D
levels, while lean body mass, participation in sports and summer and autumn seasons were positively
associated. These findings suggest that lifestyle interventions can improve vitamin D status.
Inadequate levels of vitamin D is a risk factor for osteoporotic fractures (174;178), and treatment with
vitamin D and calcium reduces fracture risk in studies consisting of both men and women (179;180),
although specific data for men are lacking. In elderly men, vitamin D has been positively associated
with hip and spine BMD (181), and vitamin D levels below 50 nmol/l have been associated with
increased bone loss at the hip (182). Vitamin D insufficiency has been shown to increase the risk of hip
fractures (183). In the NHANES III, vitamin D and BMD were associated irrespective of sex, race and
age (184), but recent data from the same study showed an effect of vitamin D deficiency on BMD
primarily in white men (185).
Several factors influence the level of vitamin D. The MrOS (162), along with several other studies, has
shown inverse associations between vitamin D and estimates of fat mass, e.g.. waist-hip ratio and total
body fat mass measured by DXA (162;186-188). A closer association between vitamin D and visceral
33
fat mass than to subcutaneous fat mass has been found, possibly linking vitamin D to the metabolic
syndrome (189).
The definition of vitamin D inadequacy has been debated. Vitamin D deficiency was defined as levels
of 25-OHD below 50 nmol/l and insufficiency as levels lower than 75 nmol/l (177). Perhaps due to the
late consensus on the recommendable levels of vitamin D, currently advocated intake of vitamin D is
insufficient for the prevention of vitamin D deficiency (190). In line with that, using vitamin
supplements including vitamin D did not result in sufficient vitamin D levels in 1606 men in the MrOS
(162).
13j. FALLS
Falls are particularly common in the elderly as more than one in three aged more than 65 years
experience a fall every year and 5-6% of falls cause fractures (191). In a Swedish 12-year register-based
study on injuries including 29,000 fractures, 53% and 80% of fractures in persons aged >50 years and
>75 years, respectively, were judged to be caused by low-energy trauma (192). Men experience fewer
falls than women but are twice as likely to die from the incident (193). In Iceland, more than 70% of
hip and forearm fractures in men were caused by low-energy trauma including falls (55), while in the
MrOS 60% of all vertebral fractures were caused by falls (18).
While the direct consequences of falls include fractures and head injuries, falls also cause long-lasting
disabilities due not only to the fracture but also to the fear of another fall (194). A multidisciplinary
approach to the prevention of falls is generally recommended (195), and recent Cochrane reviews have
supported the use of multidisciplinary methods among elderly residents of nursing care facilities and
hospitals as well as elderly, community-dwelling persons (196;197).
Vitamin D may not only improve BMD but also non-bone functions including muscle strength,
balance and the risk of falls. Vitamin D deficiency causes myopathy whereas increasing the level of
vitamin D improves muscle strength and function as well as balance (191;198). A meta-analysis based
34
of eight randomized controlled trials including 2416 persons (19% men) by Bischoff-Ferrari et al. (199)
found that supplementation with more than 700 IU of vitamin D reduced falls by 19%. A subgroup
analysis on the effect of vitamin D in men was not reported.
Falls were also associated with an increased risk of fractures in the SOMA study. The deleterious effect
of falls appeared to be graded, with increasing numbers of falls being followed by higher risk of fracture
(HR (95%CI): one fall/year: 1.47 (1.00-2.17), 2-4 falls/year: 2.10 (1.37-3.22), >4 falls/year (2.34 (1.083.07)). In addition, the impact was most pronounced in those with a normal BMI (<25kg/m2) and in
the oldest participants (men aged >70 years).
No sex-specific intervention studies on fall prevention have been published and specific
recommendations for the prevention of falls in men are therefore not available. There are no data
indicating a sex-related effect of vitamin D, however. As in women, fall prevention in men relies on
multidisciplinary input as well as optimization of vitamin D status and eradication of all other factors
imposing a higher risk of falls.
13k. SMOKING AND ALCOHOL
Alcohol consumption and tobacco use are higher in men than in women. In a meta-analysis of the
consequences of smoking on BMD and fracture risk, smoking reduced BMD and increased fracture
risk in women and men, with higher risks for all except hip fractures observed in men (200).
Interestingly, only 23% of the effect of smoking on fracture risk was explained by detrimental effects of
areal BMD. In Denmark, smoking was associated with increased risk of a hip fracture in both men and
women, and 19% of all hip fractures may be attributed to the effect of smoking (19). Conversely,
results from the SOMA study indicated no significant association between smoking and fracture risk
(HR (95%CI): 1.22 (0.87-1.72)), however, this may be related to the relatively low number of events.
Smokers with low weight or more than 20 pack-years of smoking seemed to be particularly exposed to
35
the deleterious effects of smoking with regard to bone (170), and BMD remained low in former
smokers participating in the MINOS (201).
In a review including risk factors for osteoporosis, alcohol intake exceeding 3-4 units per day was found
to increase fracture risk by about 2.0 even though BMD was unvarying (202), implying harmful effects
of alcohol on non-bone related factors, i.e. falls. In the EPOS, alcohol had no independent relation
with fracture risk (203) and some reviews have considered the evidence of an alcohol effect on bone
health in men to be weak (153;202). Kanis et al. (204) found alcohol intake to be associated with the
risk of any fracture, osteoporotic fracture and hip fracture independent of BMD and gender, although a
trend towards larger effects in men was observed. Danish data including almost 18,000 men and 14,000
women associated consumption of more than 27 units of alcohol per week with an increased risk of hip
fracture (28-41 units/week: RR (95%CI): 1.75 (1.06-2.89) and >70 units/week: 5.28 (2.60-10.70)) in
men, whereas in women no substantial effect was found after adjustment for confounders (205).
Both smoking and excessive consumption of alcohol are thus associated with higher fracture risk in
men and women. Not all studies agree with regards to the effect of alcohol, although differences in
study design, population and size could account for the divergence.
13l. COMMON DISEASES ASSOCIATED WITH OSTEOPOROSIS AND FRACTURES
Type 2 diabetes (T2DM) has been associated with an increased fracture risk despite reports of increased
BMD in diabetics (160;206). The reason for this is unknown. Lower bone strength compared to body
weight could explain part of the observation (207). Also, falls are more prevalent among T2DM
patients, possibly due to reduced peripheral nerve function and decreased renal function, weight loss or
worsening of vision as well as intensive treatment with insulin (208). Recently, the number of
medications – but not the individual medication – prescribed to persons with diabetes was related to
the risk of falls (209). Presently, T2DM is not acknowledged as a risk factor for osteoporotic fractures
by the Danish Medicines Agency. The data are consistent and point to the need for greater attention to
36
this patient group, but further studies are required to investigate the mechanism causing the detrimental
effect of T2DM on bone status.
Since cardiovascular disease (CVD) is more prevalent in men than women, CVD could be important in
the explanation of sex differences in BMD and fracture incidence. Few studies have evaluated the
impact of CVD on fracture risk. Data from Sweden found incidence of hip fractures of 12.6/1000 in
patients with heart failure or stroke and 6.6/1000 in patients with ischemic heart disease compared to
1.2/1000 in the population without any of these diagnoses. Due to increased risk of fracture in cotwins not diagnosed with IHD, Sennerby et al. (210) suggested a genetic relation between CVD and hip
fractures. Men belonging to the lowest quartile of BMD in the spine and forearm experienced a twofold increased risk of CVD while those with aortic calcifications independently of medication,
comorbidities, falls, BMD, BMI and age had a two-three fold increase in fracture risk during the 7.5
years follow-up of the French MINOS, suggesting that the cardiovascular and skeletal systems could
interact in presently unknown ways (211;212). In a Danish register-study, hypertension, stroke and
myocardial infarction were associated with a slight increase in fracture risk in the first 3 years after the
event (213), whereas Hippisley-Cox and Coupland (160) showed that CVD was associated with an
increase in osteoporotic fractures in both men and women independent of smoking, T2DM, BMI and
several other potential confounders.
Certain lifestyle factors, including smoking, are central to the pathogenesis of CVD and osteoporosis
and also contribute to the incidence of other diseases, e.g. pulmonary disorders.
Chronic obstructive pulmonary disease (COPD) and asthma have recently been associated with bone
loss in both spine and hip as well as an increased fracture risk in the MrOS. The risk of osteoporosis
was increased in those treated with inhalation (OR (95%CI): 2.05 (1.27-3.31)) or oral steroids (OR: 2.13
(1.15-3.93)). Vertebral and non-vertebral fracture risk was increased by factors of 2.6 and 1.4 in
participants with COPD or asthma, respectively (214). In UK, asthma was found to increase fracture
37
risk to a similar extent in both women and men, independent of the use of steroids (160). The cause of
the association is, however, largely unknown.
While awareness of COPD as a risk factor for osteoporosis is still evolving, rheumatoid arthritis (RA) is
acknowledged as a risk factor for osteoporosis in women. In men, the amount of information is
limited. A Norwegian study on 366 persons, including 68 men diagnosed with RA, found deleterious
effects of treatment with glucocorticoids and beneficial effects on bone loss in those on anti-resorptive
treatment. Bone loss measured by DXA occurred in patients with RA earlier in hands than in the
remaining skeleton (215), but fracture rates were not reported. During a median follow-up of almost 50
months, 2.9% of the participating 1,055 male RA patients of a Japanese study experienced either a nonvertebral or a vertebral fracture (216). The expected number of fractures in a matching non-RA
population was not reported, making the assessment of risk conferred by RA difficult. Longitudinal
studies are needed to evaluate the impact of RA on fracture risk in men.
In the SOMA study, neither type 2 diabetes nor CVD was associated with fractures. In addition, RA
had no significant impact on the occurrence of fractures, whereas pulmonary disease increased the risk
of an osteoporotic fracture (HR: 1.95 (1.05-3.62)). Interestingly, other diseases or symptoms were
related to an increased fracture risk. Erectile dysfunction was associated with both any fracture and
osteoporotic fracture (HR: 1.39 (1.04-1.84) and 2.04 (1.31-3.18), respectively). In addition, report of
urinary frequency was associated with any fracture (HR: 2.05 (1.25-3.36)) but not osteoporotic fracture.
These results are surprising but also difficult to explain. Both urinary frequency and erectile dysfunction
are most likely not the cause of fracture but rather markers of frailty and underlining disease.
Although a substantial number of common comorbidities are known to increase fracture risk in men,
information on the mechanism increasing the fracture risk is not available. In theory, illness may
deteriorate BMD or the bone structure directly or increase fracture risk indirectly through adverse
effects of the concomitant medication, reduced level of physical activity, tendency to falls or increased
frailty.
38
13m. MEDICATION
Numerous compounds have been related to osteoporotic fractures in men including loop diuretics,
selective-serotonin reuptake inhibitors (SSRI) and glucocorticoids (GC) (217).
13m. Glucocorticoids
GCs suppress bone formation by increasing osteoblast apoptosis and bone resorption, reducing
intestinal calcium absorption and increasing renal calcium excretion, and disturbing vitamin D
metabolism, thus increasing the risk of vertebral and hip fractures in men and women (17). Inhaled
GCs, however, had no impact on hip (218) or any other fracture (219). Danish registry data on men
and women suggested an impact of oral steroids on any fracture risk at levels exceeding 7.5 mg of
prednisolone (or equivalent), while no effect of topical applications were found (220). GC appears to
affect fracture risk equally in men and women, making GC important as a risk factor irrespective of
gender.
13m. SSRI and anti-epileptic drugs
SSRIs were associated with decreased BMD and increased risk of osteoporotic fracture in large,
prospective studies and in a Danish registry-based study in men (217;221;222). SSRIs are consistently
associated with osteoporotic fractures (222), and recent data have suggested a biological mechanism for
the deleterious effect of SSRI on bone (223). In addition, SSRIs appear to increase the risk of falling in
men and women (221).
Future studies should evaluate the effects of antidepressants including SSRI as well as tricyclic
antidepressants on bone health as well as falling in men, and whether these effects are independent of
the underlying disorder, e.g. depression, anxiety etc.
13m. Anti-epileptics
Vestergaard et al. (224) showed that epilepsy doubled the risk of fracture, with higher risks in those
treated with phenytoin. Once seizure-related fractures (one third of events) were removed, however,
39
the effect was only borderline significant. In a US study on 81 young men, anti-epileptic drugs (AED)
were associated with femoral neck bone loss (225), whereas in the MrOS only non-enzyme-inducing
AED were associated with increased bone loss (226).
14. PREVENTION
The proportion of persons ultimately fracturing is larger among those with osteoporosis, but the
absolute number of fractures is higher in the non-osteoporotic population. Preventive strategies could
either focus on a reduction of fractures in persons at high risk or aim at decreasing the risk in the total
population.
Reducing the prevalence of risk factors in the general population would lower the total risk and lessen
the societal burden of fractures, but the change may have a larger impact on high-risk individuals. On
the other hand, it could be easier to reduce exposure in high-risk individuals. In order to be effective,
however, these exposures would have to be both causal and reversible which may not be the case, e.g.
age and genetics. International guidelines for osteoporotic fracture prevention have been developed,
but the likely impact of population-based guidance should be evaluated before implementation.
The prevention of osteoporosis and osteoporotic fractures starts early, as maximising PBM of the
individual appears to reduce the overall fracture risk in later life. Physical activity, smoking and adequate
nutrition including vitamin D are all relevant to PBM and have to be addressed if bone health is to be
improved. Searching for high-risk patients seems of minor importance in a young population, although
fractures in early life can be due to underlying bone-related diseases, e.g. osteogenesis imperfecta, that
require special intervention.
In adults, both population-based prevention and risk reduction in high-risk individuals appears to be
sensible ways of improving bone health. In high-risk patients, reducing exposure or protection against
the effects of an exposure would be possible, whereas preventive measures on a public health level may
reduce the overall impact of osteoporotic fractures on society. In Denmark, case-finding strategies are
40
recommended rather than population screenings (227). This approach, however, seems to divert
resources to low-risk patients and provides little coverage of high-risk man and women. Thus, a
substantial number of Danish man and women at low risk of a fracture as assessed by the WHO
fracture algorithm FRAX have been evaluated with DXA, indicating a need for optimization of the use
of DXA (228) (and Paper IV).
No studies on community screening for osteoporosis have yet been published, but the UK Screening of
Older Women for Prevention of Fracture Study (11.000 women aged 70-85 years)
(http://www.scoopstudy.ac.uk) and the Danish ROSE Study (women aged 65-80 years)
(www.interreg4a.dk/wm313724) will provide information on the usefulness of screening in women.
The present high-risk preventive strategy comes with some advantages. Assessing the individual’s risk
factors allows for information appropriate for that person specifically, and interference with those at
minor risk is avoided. In addition, it may be easier for patients and medical staff to deal with the
specific term osteoporosis rather than an absolute fracture risk. Greater focus on high-risk patients could
make long-term treatment more acceptable as the risks due to the medication would be outweighed by
the benefit of the intervention.
For clinical decision-making, identification of men at high risk of fracture presently includes measuring
BMD, as the diagnosis of osteoporosis is based on a T-score. Though significantly related to the risk of
a fracture, the diagnosis osteoporosis could be perceived as a risk factor for fracture. Osteoporosis captures
some but not all of the risk of a fracture as is the case for falls, TCAs (229) etc. The proportion of
individuals identified as osteoporotic on the basis of a T-score of -2.5 or less and who eventually
experience a fracture was 21% in the Rotterdam study (27), clearly making way for new approaches to
the management of osteoporosis including development of other surrogate markers for fracture. Future
indicators of fracture risk could be the strength of the femoral neck or vertebrae as estimated by
QCT/finite element analysis, bone structure as measured by HR-pQCT or algorithms based on
anthropometrics and medical history (see below) (131).
41
14a. FRACTURE PREDICTION
The WHO fracture risk algorithm, FRAX©, allows for calculation of the 10-year absolute fracture risk
in men and women (21). FRAX includes a number of risk factors as well as either BMI or BMD (Table
3). FRAX could prove useful in the selection of individuals eventually fracturing, particularly in men as
the sensitivity of DXA to detect those who will experience a fracture is low. The algorithm was
validated in approximately 230,000 persons, but only a few hundreds of those were men (21). Neither
dose-response relationships nor dynamic interactions between the clinical risk factors were accounted
for. In addition, clinical trials on pharmacological treatment of osteoporosis have been performed on
the basis of T-scores or factures rather than absolute risk estimates. Though not designed specifically
for formal testing of FRAX estimates, recent reports on placebo-controlled, randomized clinical trials
of clodronate (230) and bazedoxifene (231) demonstrated the ability of FRAX to identify women at
high risk of fracture who would respond to treatment. Similar studies are needed in men. At the
moment, clinical trials have provided no evidence supporting the usefulness of FRAX in men.
Two other risk algorithms have recently emerged. Compared to FRAX, the Australian Garvan (232)
algorithm incorporates fewer risk factors and the UK QFractureScore (160) substantially more risk
factors (Table 3). The algorithms were derived from rather diverging study populations, thus the
Garvan is based on results on the DOES, while the QFractureScore was constructed on the basis of
huge datasets from general practitioners in the UK. While FRAX and the Garvan algorithm gave
comparable results in women, the ability of FRAX (US version) to predict fractures in men was shown
to be inferior to the Garvan algorithm (232). Compared to FRAX, the QFractureScore appears to
provide better discrimination of fracture patients (160). In addition, the QFractureScore is applicable to
a larger age range and accommodates several factors not present in FRAX such as type 2 diabetes and
falls. This algorithm has been validated in >400,000 men and women in the UK. On the other hand,
42
FRAX is based on international data and additions of more risk factors may not necessarily prove
sensible in a clinical setting. Validation of the three algorithms is needed, particularly in men.
15. CONSEQUENCES OF FRACTURES
Men are less likely than women to return to their own home after hospital discharge following hip
fracture (233). One in four male fracture patients who were previously self-sufficient became
dependent on others for help at home, while another 50% of those who were previously independent
were admitted to residential homes or institutionalized care (234;235). In addition, men are less likely
than women to have returned to their previous independent living one year after hip fracture (233).
The mortality associated with hip fractures is higher in men than in women. In a recent Danish study,
Kannegaard et al. found a 1-year mortality among hip fracture patients (236). Comparable results have
been found in a number of countries (14;68;237;238). The increased mortality after a hip fracture –
particularly in men – has been confirmed in a meta-analysis (239). Men face twice the risk of dying on
admission due to a hip fracture compared with women (240), and their 1-year mortality remains higher
in men after a second hip fracture (53). In contrast, the Canadian Multicentre Osteoporosis Study
(CaMOS) found no sex-specific difference in mortality after hip fracture, quite possibly due to a limited
number of events (n=85)(67).
In men with hip fracture, the number of co-morbid conditions is higher than in women. Nonetheless,
the excess mortality remained after adjustment for several comorbidities and medications, suggesting
that other factors play a role (236). Unfortunately, there is at present no convincing explanation for the
increased mortality following fractures observed in men.
15a. CLINICAL CARE
In the CaMOS, 20% of the male study population reported a previous fragility fracture at study
inclusion in 1996-7; however, only 2.3% had been diagnosed with osteoporosis. After 5 years of study
43
participation, the fraction of fracture patients diagnosed with osteoporosis increased to approximately
10%, and less than 40% of hip and 10% of the wrist fracture patients received treatment after 5 years
(16). Bone densitometry was used four times more often in women compared to men in an Australian
study (241), even though the life-time risk of a fracture in women and men over 50 years old is
estimated to be 44% and 27%, respectively (242), and 39% of all fractures occur in men (3). Although
the use of anti-osteoporotic treatment has increased in Denmark (243), the disease remains
underdiagnosed and inadequately treated, particularly in men (7).
Using baseline self-report data on diagnosis of osteoporosis, use of osteoporosis medication and
previous bone mass assessment in a population-based sample of 4,975 Danish men aged 60-74 years,
we found that 2.7% of the men had been assessed by DXA (Paper IV). Bone mass assessment was
used more often among those with risk factors or a high FRAX score; however, DXA had been
performed in only 10% of men with at least three FRAX risk factors as compared to 36% of elderly
Danish women (228). Furthermore, 10% of men not assessed by DXA had a 10-year major
osteoporotic fracture risk in excess of 20%. Conversely, 32% of the bone assessments in the same
study had been performed in men without clinical risk factors as defined by the FRAX algorithm.
These results suggest that the strategy used to prevent fractures in men needs to be re-evaluated, and
further studies focusing on the causes of the observed sex-related differences are needed.
16. PHARMACOLOGICAL THERAPHY
When men and women are diagnosed with osteoporosis, the aim of the treatment is fracture reduction.
Randomized, double-blinded, placebo-controlled studies on fracture reduction are needed to provide
necessary information on efficacy and adverse effects of the treatment. The European Medicines
Agency (EMEA) has licensed drugs for the treatment of male osteoporosis on the basis of fracture
reduction in women and significant increase of BMD in men (www.emea.europa.eu) if this increase was
44
comparable with that observed in women. Only a few studies have investigated the treatment of
osteoporosis in men.
16a. VITAMIN D
Several studies have investigated the impact of vitamin D supplementation on bone health and fracture
risk. However, meta-analyses have produced varying results. Bischoff-Ferrari et al. (244) found that
17.5-20 µg vitamin D reduced fracture risk, whereas Boonen et al. (245) reported that vitamin D was
effective in reducing fracture risk only if administered with calcium. Recently, the need for a
combination of vitamin D and calcium for fracture reduction was supported by an even larger metaanalysis including 65,000 individuals (14% men) (2). Interestingly, in the latter study, there was no
indication of an interaction between sex and response to treatment, suggesting that vitamin D plus
calcium may be as effective in men as it is in women.
16b. BISPHOSPHONATES
Orwoll et al. (246) showed significantly reduced risk of vertebral fractures and increased BMD of the
spine, femoral neck and whole body in a double-blinded, placebo-controlled, randomized trial on
alendronate treatment of 241 men with mean age of 63 years, of whom one-third were hypogonadal
(defined as low total testosterone). Positive effects on BMD were also shown in studies on alendronate
(247;248), risedronate (249) and ibandronate (250). Ibandronate also increased BMD of the lumbar
spine and femoral neck in heart transplant patients (251), whereas zoledronic acid reduced the
incidence of all fractures including vertebral fractures and increased BMD at several sites in a 3-year
study with male and female hip fracture patients (252). Contrary to all other treatment options,
zoledronic acid also reduced mortality in hip fracture patients (252). Together, these studies
demonstrated similar effects of bisphosphonates on bone loss and osteoporosis in men as in women.
None of the studies suggested sex-specific adverse effects.
45
16c. PTH
Although terminated prematurely due to development of osteosarcomas in rats, a study including 437
elderly men on teriparatide for the treatment of male osteoporosis showed beneficial effects on spinal
and femoral neck BMD. In addition, the effect was present irrespective of gonadal status (253). Similar
results are available from a smaller study of longer duration that in addition showed a reduced
incidence of vertebral fractures in the actively treated group of participants (254). As in women,
bisphosphonates are effective in maintaining bone mass after anabolic treatment with teriparatide (255).
A combination of teriparatide and bisphosphonates (alendronate) in men appears to reduce the effect
of teriparatide (256) and is not recommended. Data on the effect of parathyroid hormone (Preotact) on
male osteoporosis are not available, but the function of parathyroid hormone is not supposed to be
sex-specific. As is the case for bisphosphonates, there appears not to be any sex-specific adverse effects
of PTH – there is, however, no clinical data on 1-84 PTH in men.
16d. TESTOSTERONE AND SERM
A few, extremely divergent studies have evaluated the effect of testosterone on BMD whereas data on
fracture prevention is absent. Behre et al. (257) found a 20% increase in trabecular bone mass as
measured by QCT in the first year of testosterone treatment of 72 hypogonadal men. The results are
difficult to extrapolate, however, as various formulations of testosterone were used, no control group
was included and there were several different causes of the hypogonadism.
Testosterone treatment has been tested for prevention of bone loss in hypogonadal men and beneficial
effects of testosterone have been shown on measures of BMD, although findings are not consistent.
46
Snyder et al. (258) found increased BMD of the spine in a study on 100 elderly men. Positive effects on
BMD have also been reported in other studies (259;260).
Selective oestrogen receptor modulators (SERM) imitate the effects of oestrogens on bone. In a
randomized, placebo-controlled trial on the effects of raloxifene on bone turnover markers, treatment
was associated with decreased bone resorption (urinary NTX) in 50 elderly men with low oestradiol
level (261). Unfortunately, BMD was not evaluated in the study. Confirmatory studies have not been
published. SERMs seem to be associated with an increased risk of thromboembolic events in men with
prostate cancer, an adverse effect similar to that observed in women (261).
16e. OTHER DRUGS
There are no data to suggest sex-specific effects of strontium on bone. Data on strontium ranelate for
treatment of male osteoporosis is currently not offered. Although calcitonin has been tested in men, the
trials were small, of short duration and lacked data on fractures. Trovas et al. (262) found significant
increases in spinal BMD and reduced bone resorption markers during one year of treatment, but no
effects were noted in the hip or femoral neck. There appears not to be any sex-specific adverse effects
of calcitonin.
16e. Special case: glucocorticoid-induced osteoporosis
Several bisphosphonates have proven useful in the treatment or prevention of osteoporosis in men
currently on or beginning glucocorticoid-treatment including alendronate (263), risedronate (264),
ibandronate (265) and zoledronic acid (266). Zoledronoic acid and teriparatide appear superior to
alendronate in maintaining BMD in the lumbar spine (266-268). Most studies on the treatment of
glucocorticoid-induced osteoporosis (GIO) include both sexes since there appear not to be any
significant sex-difference in GIO.
16e. Special case: prostate cancer
47
Bisphosphonates have proven effective in maintaining or increasing BMD at the beginning of or during
anti-androgen treatment (ADT) in a number of trials of various sizes and durations (269-273). More
recently, denosumab, a human monoclonal antibody against receptor activator of nuclear factor kappaB
ligand, decreased the risk of a vertebral fracture in men undergoing ADT due to non-metastatic
prostate cancer and increased BMD in the lumbar spine and total hip in a study including 1,468 men
(274).
SERMs have been evaluated as a preventive measure in men with prostate cancer receiving ADT. A
combination of SERM and ADT that increased oestrogen but maintained a suppressed level of
androgens could prove effective in preventing deleterious effects of ADT on bone. In a small study,
raloxifene increased BMD in the hip (275), whereas toremifene, a SERM, reduced the risk of vertebral
fractures and increased BMD in a 2-year study on 646 prostate cancer patients undergoing ADT (276).
16f. TREATMENT IN GENERAL
Bisphosphonates have a documented effect on male osteoporosis including cases due to hypogonadism
and bone protection during ADT and GC treatment. The PTH-analogue teriparatide is effective in
treating male osteoporosis and GIO and appears more potent in treating GIO than bisphosphonates in
men. Lately, denosumab has become an option for treating bone loss in prostate cancer patients
receiving ADT. A head-to-head study with bisphosphonates for this particular group of patients has
not been published and the choice between these drugs is thus not clear.
Calcitonin, SERMs and testosterone have not been shown to reduce fracture risk in men and appear
irrelevant to the treatment of male osteoporosis. Nevertheless, testosterone is available for substitution
in hypogonadism and could improve bone status directly or via the peripheral conversion to oestradiol
by aromatase or by increasing muscle mass.
48
Ongoing studies are investigating the effects of zoledronic acid and denosumab on bone status in men
and evaluating the 24-month effect of teriparatide in men and women as assessed by HR-pQCT
(www.clinicaltrials.gov).
17. CONCLUSION
In a population-based study of otherwise healthy young Danish men, the prevalence of vitamin D
insufficiency was found to be high during winter, affecting almost half of the study population. The
levels of vitamin D were inversely associated with BMD, suggesting that inadequate vitamin D could be
detrimental to peak bone mass. Whether this inadequacy is a marker of future low BMD and increased
fracture risk remains to be answered.
Low BMD and osteoporosis are predictive of fractures in men as in women. Based on a populationbased sample and Danish reference values, 10% of elderly Danish men aged 60-74 had osteoporosis.
Using either Danish or NHANES reference values resulted in the same overall prevalence of
osteoporosis, although the estimates differed substantially between the different anatomical regions.
Estimating the prevalence of osteoporosis on the basis of a single skeletal site will, therefore, provide
very different results.
Although osteoporosis is a common disorder of elderly men and women, only very few Danish men at
high risk of fractures had been evaluated with DXA. The majority of those who reported a previous
DXA were at high risk of a fracture, but almost one-third of the DXAs had been performed in
individuals at low risk, suggesting that the pattern of referral for DXA was inappropriate. The use of
DXA may also allow for assessment of vertebral fractures. In this study, the prevalence of vertebral
fracture was 6% as assessed by VFA. Although one in eight of the participants had either osteoporosis
or a vertebral fracture, however, less than one per cent reported osteoporosis-specific treatment.
49
Although male osteoporosis has been investigated to a lesser degree than female osteoporosis, there is a
current body of knowledge about the causes of osteoporosis in men, covering both reversible and
irreversible factors. Thus, insufficient levels of sex steroids and quite possibly increased levels of sex
hormone-binding globulin confer a higher risk of osteoporosis and fractures. Heritability also
influences the occurrence of fractures in men and women, and common variations in several genes
have been shown to influence fracture risk. Large scale genetic studies have provided limited
information on the origin of fractures and there is no indication of sex-specific differences in the
effects of genetics. The information retrieved from genetic studies nevertheless explains a minor part of
the fracture risk compared to age, BMI, falls, previous fractures along with numerous other factors
known to increase the prevalence of osteoporosis and fracture risk in men.
In a population-based sample of elderly men followed over five years through the use of registers, it
was found that family history of a hip fracture, weight loss, falls, erectile dysfunction and urinary
frequency may contribute to information about fracture risk in men. Some of these factors are related
to osteoporosis, while others are probably markers of frailty or underlining disease.
Recently developed algorithms allow calculation of absolute fracture risk and may emerge as a more
prudent method for detecting patients likely to benefit from treatment. However, there is a need for
further validation of these algorithms in men. In addition, while there are several treatment options
approved for use in men and even more presently under investigation, none of them has been shown
to be effective in men on the basis of an estimated absolute fracture risk.
Although treatment is available, male osteoporosis remains underdiagnosed and undertreated. DXA is
used more commonly in women. Even though men have frequent fractures and several well-known
risk factors, they are less likely to start specific anti-osteoporotic treatments.
50
18. PERSPECTIVES
There are major unanswered questions. First, are the available algorithms for fracture prediction valid
in Danish men? The validity of the Garvan algorithm and the QFractureScore can be investigated using
the SOMA study, whereas other studies with longer follow-up time are needed for evaluation of FRAX
in Danish men. Second, what are the causes of the increased mortality observed in men following
fractures? As with the difference in longevity between men and women, there is no biological
explanation for the increased mortality in men with fractures. Increased levels of comorbidity may
explain parts of the difference. Evaluating monozygotic twins disconcordant with regard to fracture
could potentially contribute information due to their shared milieu and genetics. Third, is screening for
osteoporosis in men justified? And is it prudent to treat on the basis of a high fracture risk rather than a
T-score? While at least two studies on population screening in women are undergoing, there are no
such investigations in men. A large-scale study investigating the feasibility of male population screening
is certainly needed as osteoporosis defined on the basis of a T-score appears to be inadequate for men.
51
19. APPENDICES
Table 1. Randomized studies on pharmaceutical treatment of osteoporosis in men (vitamin D
excluded).
Bisphosphonates
Alendronate
Study
Participants n
Age
Design
Diagnosis
Duration
Endpoint
2 years
BMD
Conclusion
[%men if mixed
cohort]
Orwoll (246)
241
63
DB RCT
(10mg(po)/day)
Low T-score(-2)
BMDspine:+7.1%
and/or previous
BMDneck:+2.5 %
fracture
BMDwb:+2%
Vfx (treated: 0.8% vs.
PBO: 7.1%, p=0.02)
Ringe (247)
134
Randomized,
T-score<-2.5 in
open-label
the lumbar
2 year
BMD
6 mo
BMD
spine
Hwang (277)
46
Randomized,
(70mg(po)/week)
Greenspan (269)
open-label
112
(70mg(po)/week)
Gonnelli (248)
T-score<-2.5
77
BMDneck: +2.7%
DB RCT,
Non-metastatic
partial cross
prostate cancer,
over trial
ADT
57
BMDspine:+5.5%
T-score<-2.5
2 year
BMD
BMDspine:+6.7%.
BMDneck:+3.2%
3 year
(10mg(po)/day)
BMD,
BMDspine:+8.8%.
QUS
BMDneck:+4.2%
BUA:+3.8%
Saag (263)
477 [30% men]
DB RCT
(10mg(po)/day vs.
Glucocorticoids
48 weeks
BMD
>=7.5mg
BMDspine: +2.9%
(10mg) vs. PBO:-0.4%
5mg(po)/day vs.
BMDneck:+1.0% vs.
PBO)
PBO:-1.2%
de Nijs (278)
201 [38% men]
DB RCT
Starting
(10mg(po)/day vs
glucocorticoids
alfacalcidol
>=7.5mg
18 mo
BMD
BMDspine: +2.1% vs,
alfa: -1.9%
(1mg(po)/day)
Risedronate
Boonen (249)
284
DB RCT
T-scores: <-2.5
2 year
BMD,
BMDspine:+4.5%
52
(35mg(po)/week)
spine or <-2 hip
sec. Vfx
(male reference)
Wallach (264)
518 [%men]
DB RCT
??
Vfx: treated 4.9% vs.
7.7%, NS
1 year
(2.5mg(po)/day vs.
BMD,
BMDspine:+1.9% (5mg)
sec. VFx
vs. PBO:-1.0%
5mg(po)/day vs.
VFx: RR: 0.30
PBO)
Ibandronat
Orwoll (250) (150mg
132
DB RCT
(po)/mo)
Ringe (265)
104 [47% men]
T-scores BMD
1 year
BMD
BMDspine:+3.5% vs.
spine or
PBO:0.9%
BMDneck <-2,
BMDneck:1.2% vs.
>30years
PBO:-0.2%
controlled,
2 year
2 year
BMDspine:+11.9% vs.
(2mg(iv)/3mo vs.
prospective,
treatment with
alfa:+2.2%
1mg alfacalcidol)
open-label,
>=7.5mg
BMDneck:+4.7% vs.
parallel-group
corticosteroid
alfa:+1.3%
study
Fahrleitner-Pammer
35
RCT
(251) (2mg(iv)/3mo)
Cardiac
1 year
transplant
BMD,
VFx: 13% vs. PBO: 53%.
VFx
BMDspine/hip: no
change in treated.
BMDspine/hip: -25%
and -23% in PBO
Pamidronate
Smith (271) (60mg
47
(sc)/12 weeks)
Randomized,
ADT + non-
48 we
BMD
BMDspine/hip: no
open-label
metastatic
change in treated.
prostate
BMDspine/hip: -3.3%
cancer.
and -1.8% in PBO
Zoledronic acid
Orwoll (279)
302
DB RCT
883 [%men]
DB RCT
2 years
BMD
BMDspine:
1 year
BMD
BMDspine:
(5mg/iv)/year vs.
70mg
alendronate(po)/week
Reid (266)
Treatment:
(5mg(iv)/year vs 5mg
>3mo
Treatment: Zol:+4.1% vs.
risedronate(po)/day)
glucocorticoid
Ris:+2.7%
Prevention: <3
Prevention: Zol:+2.6%
mo
vs. Ris:+2.0%
53
glucocorticoid
(both non-inferior and
superior)
Lyles (252) (5mg
2127 [24% men)
75
DB RCT
Hip fracture
3 years
Fx, BMD
(iv)/year)
Fx: 8.6% PBO: 13.9%.
HR: 0.65 (95%CI:0.500.84)
BMDhip:+5.5% vs.
PBO:-0.9%
BMDneck:+3.6% vs.
PBO:-0.7%
Satoh (270)
40
ADT + non-
(4mg(iv)/year)
1 year
BMD
metastatic
BMDspine:+5.1% vs.
controls: -4.6%.
prostate
cancer.
Smith (272)
106
DB RCT
(4mg(iv)/3 mo)
ADT + non-
1 year
BMD
BMDspine:+5.6%
1 year
BMD
BMDspine:+4.0% vs.
metastatic
prostate
cancer.
Michaelson (273)
40
RCT
(4mg(iv)/1year)
Ongoing ADT,
non-metastatic
PBO: -3.1%.
prostate cancer
BMDhip:+0.9% vs.
PBO:-1.9%)
SERMs
Raloxifene
Smith (275)
44
70
(60mg(po)/day)
RCT, open-
Ongoing ADT,
BMDspine:+1.0% vs.
label
non-metastatic
PBO: -1.0% (NS).
prostate cancer
BMDhip:+1.1% vs.
PBO:-2.6%)
Toremifene
Smith (1)
1,284
DB RCT
(80mg(po)/day)
Ongoing ADT,
2 year
Fracture
RR 50%.
non-metastatic
risk,
Only epub available as of
prostate cancer
BMD
August 2010
BMD
BMDspine:+13.5%
PTH analogs
Teriparatide (1-34PTH)
Kurland (254)
(400IU(sc)1-34PTH
23
50
DB RCT
z-scores <-2 of
t-scores <-2.5
18mo
BMDneck:+2.9%
54
in spine or
femoral neck
(male
reference)
Orwoll (253)
T-scores <-2
Median
(20microg (sc),
in spine or hip
11 mo
(20microg), +9.0%
40microg (sc) or PBO
(male
(due to
(40microg)
(sc))
reference)
cancer in
BMDneck:+1.5% and
rats)
+2.9%
Saag (267) (20microg
437
DB RCT
428
DB RCT
Min 3 mo of
(sc) vs. 10mg
>=5 mg
alen(po)/day)
glucocorticoid
Finkelstein (280)
83
58
RCT
BMDspine or
18mo
BMD
BMDspine:+5.9%
BMD,
BMDspine: +7.2% vs.
sec. VFx
alen:3.4%
VFx: 0.6% vs. alen: 6.1%
30 mo
BMD
BMDspine: alen:+11.1%,
(40microg vs.
BMDneck <-2
alen+teriparatide:+18.0%,
40microg+10mg
(male
teri:+25.8%
alen(po)/day vs.10mg
reference)
BMDneck: alen:+3.2%,
alen(po)/day)
Langdahl (268)
alen+teri:6.2%, teri: 9.7%
83 [%men]
DB RCT
(20microg (sc) vs.
GIO (T-
18mo
BMD
score<-2.5)
BMDspine: 7.3% vs.
3.7%
10mg
alendronate(po)/day)
Calcitonin
Trovas (262) (200IU
28
52
DB RCT
T-score<-2.5
1 year
calcitonin (nasal)/day
BMD,
BMDspine:+7.1% vs.
BTM
PBO:+2.4%
BMD,
Vfx: treated: 1.5% vs.
sec. Vfx
PBO: 3.9%, p=0.006)
vs. PBO)
Denosumab
Smith (274;281)
1,486
DB RCT
(denosumab vs. PBO)
ADT + non-
3 years
metastatic
prostate
BMDspine:+7.9%
cancer. Age
BMBhip:+5.7%
(>70years), Tscore<-1 or
osteoporotic
fracture
Testosterone
Snyder (258)
108
73
DB RCT
T<1SD and
3 years
BMD
BMDspine:+4.2% vs.
55
(6mg(sdermal)/day)
BMDspine
PBO:+2.5%
(few received 125IU
<mean from
The lower pre-treatment
vitamin D)
young
T the larger increase
reference
Kenny (259)
77
76
RCT
Bioavailable T
1 year
BMD,
BMDneck:+0.3% vs.
(5mg(dermal)/day)
at lower limit
LBM,
PBO:-1.6%
(all 400IU vitamin D)
for adult
BFAT,
No effect BTM, increased
*34% drop-out
normal range
BTM
LBM and decreased
BFAT
Svartberg
Emmelot-Vonk (282)
223
67
DB RCT
T<13.7nM
1 year
BMD,
6 mo
BMD,
(T undecenoate
(lower half of
body
160/day (po) or
population-
comp etc
PBO)
based
No effect on bone
distribution)
Amory (260) (T
70
71
DB RCT
T<12.1nmol/L
3 years
BMD
BMDspine:+10.2%(T),
enanthate 200mg
+9.3%(T+F) vs. PBO:
(im)/2wk+PBO or T
+1.3%
enanthate 200mg
BMDhip:+2.7%(T),
(im)/2wk +
+2.2%(T+F) vs. PBO:-
finasteride
0.2%
(5mg(po)/day) or
double-dummy) *29%
drop-out
ADT: anti-androgen treatment. BFAT: total body fat. DB RCT: Double-blinded, randomized, placebocontrolled. GIO: Glucocorticoid-induced osteoporosis. LBM: lean body mass. T: testosterone. VFx:
vertebral fracture.
56
Table 2. Recommendations regarding use of DXA.
Men
Women
Notes
Danish Bone Society(227)
Fragility fracture OR risk factor
Fragility fracture OR risk factor
National Osteoporosis
>50 years AND risk factor AND
>50 years AND risk factor AND
No need for DXA in women with fragility
Guideline Group(283)
Frax score within upper and lower
Frax score within upper and lower
fracture or men and women with a high risk
assessment threshold
assessment threshold
as assessed by the Frax score
National Osteoporosis
>70 years OR >50 years + risk
>65 years OR >50 years + risk
Foundation(29)
factor OR fracture after age 50
factor OR fracture after age 50
years
years
International Society of
>70 years OR >50 years + risk
>65 years OR >50 years + risk
Clinical Densitometry (28)
factor OR fracture after age 50
factor OR fracture after age 50
years
years
>65 years AND risk factor
>65 years AND risk factor
American College of
Increased risk i.e. >70 years AND
>65 years OR >60 years AND
Risk based on risk factors rather than age
Physicians(285;286)
candidates for treatment
increased risk
defining those men that could be candidates
Canadian Osteoporosis
Society (284)
for DXA
57
Table 3. Algorithms for absolute fracture predictions.
FRAXTM 1
QFractureScoreTM 2
Garvan 3
Age
Age
Age
Sex
Sex
Sex
Fragility fracture in adulthood
Fractures since the age of 50
Falls
Falls over last 12 months
Weight and height / or BMD¤
Weight and height
BMD# or weight
More than 3 units of alcohol/week
Alcohol consumption
Smoking status
Smoking status
Use of glucocorticoids (7.5mg/3mo)
Regular use of steroid tablets
Tricyclic antidepressants
Rheumatoid arthritis
Rheumatoid arthritis
Secondary osteoporosis§ (incl. liver disease)
Chronic liver disease
Type 2 diabetes
Heart attack, angina, stroke,
transient cerebral ischemia
Asthma
Family history of a hip fracture
1
http://www.sheffield.ac.uk/FRAX/, 2 http://www.clinrisk.co.uk/qfracture/, 3
http://www.garvan.org.au/bone-fracture-risk/
¤
Femoral neck BMD or T-score(female reference), # T-score or actual BMD irrespective of site,
§
disease associated with fracture, i.e. liver disease, type I diabetes, osteogenesis imperfecta,
hypogonadism, premature menopause, chronic malnutrition, malabsorption and long-term untreated
thyrotoxicosis.
58
20. REFERENCES
Reference List
1. Oeppen J, Vaupel JW 5/10/2002 Demography. Broken limits to life expectancy. Science 296:1029-1031.
2. Christensen K, Doblhammer G, Rau R, Vaupel JW 10/3/2009 Ageing populations: the challenges ahead. Lancet
374:1196-1208.
3. Johnell O, Kanis JA 12/2006 An estimate of the worldwide prevalence and disability associated with osteoporotic
fractures. Osteoporos Int 17:1726-1733.
4. Abrahamsen B, Vestergaard P 3/2010 Declining incidence of hip fractures and the extent of use of antiosteoporotic therapy in Denmark 1997-2006. Osteoporos Int 21:373-380.
5. 6/1993 Consensus development conference: diagnosis, prophylaxis, and treatment of osteoporosis. Am J Med
94:646-650.
6. Kanis JA, Melton LJ, III, Christiansen C, Johnston CC, Khaltaev N 8/1994 The diagnosis of osteoporosis. J Bone
Miner Res 9:1137-1141.
7. Vestergaard P, Rejnmark L, Mosekilde L 2/2005 Osteoporosis is markedly underdiagnosed: a nationwide study
from Denmark. Osteoporos Int 16:134-141.
8. Sambrook P, Cooper C 6/17/2006 Osteoporosis. Lancet 367:2010-2018.
9. Center JR, Bliuc D, Nguyen TV, Eisman JA 1/24/2007 Risk of subsequent fracture after low-trauma fracture in
men and women. JAMA 297:387-394.
10. Abrahamsen B, van ST, Ariely R, Olson M, Cooper C 10/2009 Excess mortality following hip fracture: a
systematic epidemiological review. Osteoporos Int 20:1633-1650.
11. Johnell O, Kanis JA 11/2004 An estimate of the worldwide prevalence, mortality and disability associated with hip
fracture. Osteoporos Int 15:897-902.
12. Ahmed LA, Schirmer H, Bjornerem A, Emaus N, Jorgensen L, Stormer J, Joakimsen RM 2009 The gender- and
age-specific 10-year and lifetime absolute fracture risk in Tromso, Norway. Eur J Epidemiol 24:441-448.
13. Geusens P, Dinant G 2007 Integrating a gender dimension into osteoporosis and fracture risk research. Gend Med
4 Suppl B:S147-S161.
14. Kiebzak GM, Beinart GA, Perser K, Ambrose CG, Siff SJ, Heggeness MH 10/28/2002 Undertreatment of
osteoporosis in men with hip fracture. Arch Intern Med 162:2217-2222.
15. Morris CA, Cabral D, Cheng H, Katz JN, Finkelstein JS, Avorn J, Solomon DH 7/2004 Patterns of bone mineral
density testing: current guidelines, testing rates, and interventions. J Gen Intern Med 19:783-790.
16. Papaioannou A, Kennedy CC, Ioannidis G, Gao Y, Sawka AM, Goltzman D, Tenenhouse A, Pickard L, Olszynski
WP, Davison KS, Kaiser S, Josse RG, Kreiger N, Hanley DA, Prior JC, Brown JP, Anastassiades T,
Adachi JD 4/2008 The osteoporosis care gap in men with fragility fractures: the Canadian Multicentre
Osteoporosis Study. Osteoporos Int 19:581-587.
17. van Staa TP, Leufkens HG, Abenhaim L, Zhang B, Cooper C 12/2000 Oral corticosteroids and fracture risk:
relationship to daily and cumulative doses. Rheumatology (Oxford) 39:1383-1389.
18. Freitas SS, Barrett-Connor E, Ensrud KE, Fink HA, Bauer DC, Cawthon PM, Lambert LC, Orwoll ES
11/24/2007 Rate and circumstances of clinical vertebral fractures in older men. Osteoporos Int.
59
19. Hoidrup S, Prescott E, Sorensen TI, Gottschau A, Lauritzen JB, Schroll M, Gronbaek M 4/2000 Tobacco
smoking and risk of hip fracture in men and women. Int J Epidemiol 29:253-259.
20. Richards JB, Rivadeneira F, Inouye M, Pastinen TM, Soranzo N, Wilson SG, Andrew T, Falchi M, Gwilliam R,
Ahmadi KR, Valdes AM, Arp P, Whittaker P, Verlaan DJ, Jhamai M, Kumanduri V, Moorhouse M, van
Meurs JB, Hofman A, Pols HA, Hart D, Zhai G, Kato BS, Mullin BH, Zhang F, Deloukas P, Uitterlinden
AG, Spector TD 5/3/2008 Bone mineral density, osteoporosis, and osteoporotic fractures: a genomewide association study. Lancet 371:1505-1512.
21. Kanis JA, Oden A, Johnell O, Johansson H, De Laet C, Brown J, Burckhardt P, Cooper C, Christiansen C,
Cummings S, Eisman JA, Fujiwara S, Gluer C, Goltzman D, Hans D, Krieg MA, La CA, McCloskey E,
Mellstrom D, Melton LJ, III, Pols H, Reeve J, Sanders K, Schott AM, Silman A, Torgerson D, van ST,
Watts NB, Yoshimura N 8/2007 The use of clinical risk factors enhances the performance of BMD in
the prediction of hip and osteoporotic fractures in men and women. Osteoporos Int 18:1033-1046.
22. Nelson HD, Haney EM, Dana T, Bougatsos C, Chou R 7/20/2010 Screening for osteoporosis: an update for the
U.S. Preventive Services Task Force. Ann Intern Med 153:99-111.
23. Johnell O, Kanis JA, Oden A, Johansson H, De Laet C, Delmas P, Eisman JA, Fujiwara S, Kroger H, Mellstrom
D, Meunier PJ, Melton LJ, III, O'Neill T, Pols H, Reeve J, Silman A, Tenenhouse A 7/2005 Predictive
value of BMD for hip and other fractures. J Bone Miner Res 20:1185-1194.
24. Cummings SR, Cawthon PM, Ensrud KE, Cauley JA, Fink HA, Orwoll ES 10/2006 BMD and risk of hip and
nonvertebral fractures in older men: a prospective study and comparison with older women. J Bone
Miner Res 21:1550-1556.
25. De Laet CE, van der KM, Hofman A, Pols HA 12/2002 Osteoporosis in men and women: a story about bone
mineral density thresholds and hip fracture risk. J Bone Miner Res 17:2231-2236.
26. Melton LJ, III, Atkinson EJ, O'Connor MK, O'fallon WM, Riggs BL 12/1998 Bone density and fracture risk in
men. J Bone Miner Res 13:1915-1923.
27. Schuit SC, van der KM, Weel AE, De Laet CE, Burger H, Seeman E, Hofman A, Uitterlinden AG, van Leeuwen
JP, Pols HA 1/2004 Fracture incidence and association with bone mineral density in elderly men and
women: the Rotterdam Study. Bone 34:195-202.
28. International Society of Clinical Densitometry 2007 ISCD Official positions combined adult and pediatric.
29. National Osteoporosis Foundation 2009 Clinician's Guide.
30. Looker AC, Johnston CC, Jr., Wahner HW, Dunn WL, Calvo MS, Harris TB, Heyse SP, Lindsay RL 5/1995
Prevalence of low femoral bone density in older U.S. women from NHANES III. J Bone Miner Res
10:796-802.
31. Kanis JA, Johnell O, De Laet C, Jonsson B, Oden A, Ogelsby AK 7/2002 International variations in hip fracture
probabilities: implications for risk assessment. J Bone Miner Res 17:1237-1244.
32. Scholtissen S, Guillemin F, Bruyere O, Collette J, Dousset B, Kemmer C, Culot S, Cremer D, Dejardin H,
Hubermont G, Lefebvre D, Pascal-Vigneron V, Weryha G, Reginster JY 7/2009 Assessment of
determinants for osteoporosis in elderly men. Osteoporos Int 20:1157-1166.
33. Kanis JA, Johnell O, Oden A, Jonsson B, De Laet C, Dawson A 11/2000 Risk of hip fracture according to the
World Health Organization criteria for osteopenia and osteoporosis. Bone 27:585-590.
34. Donaldson LJ, Cook A, Thomson RG 9/1990 Incidence of fractures in a geographically defined population. J
Epidemiol Community Health 44:241-245.
60
35. Cooper C, Dennison EM, Leufkens HG, Bishop N, van Staa TP 12/2004 Epidemiology of childhood fractures in
Britain: a study using the general practice research database. J Bone Miner Res 19:1976-1981.
36. Singer BR, McLauchlan GJ, Robinson CM, Christie J 3/1998 Epidemiology of fractures in 15,000 adults: the
influence of age and gender. J Bone Joint Surg Br 80:243-248.
37. Khosla S, Amin S, Orwoll E 6/2008 Osteoporosis in men. Endocr Rev 29:441-464.
38. Clark EM, Ness AR, Tobias JH 2/2008 Bone fragility contributes to the risk of fracture in children, even after
moderate and severe trauma. J Bone Miner Res 23:173-179.
39. Pye SR, Tobias J, Silman AJ, Reeve J, O'Neill TW 7/2009 Childhood fractures do not predict future fractures:
results from the European Prospective Osteoporosis Study. J Bone Miner Res 24:1314-1318.
40. Kanis JA, Pitt FA 1992 Epidemiology of osteoporosis. Bone 13 Suppl 1:S7-15.
41. van Staa TP, Dennison EM, Leufkens HG, Cooper C 12/2001 Epidemiology of fractures in England and Wales.
Bone 29:517-522.
42. van Staa TP, Dennison EM, Leufkens HG, Cooper C 12/2001 Epidemiology of fractures in England and Wales.
Bone 29:517-522.
43. Mackey DC, Lui LY, Cawthon PM, Bauer DC, Nevitt MC, Cauley JA, Hillier TA, Lewis CE, Barrett-Connor E,
Cummings SR 11/28/2007 High-trauma fractures and low bone mineral density in older women and
men. JAMA 298:2381-2388.
44. Seeley DG, Browner WS, Nevitt MC, Genant HK, Scott JC, Cummings SR 12/1/1991 Which fractures are
associated with low appendicular bone mass in elderly women? The Study of Osteoporotic Fractures
Research Group. Ann Intern Med 115:837-842.
45. Stone KL, Seeley DG, Lui LY, Cauley JA, Ensrud K, Browner WS, Nevitt MC, Cummings SR 11/2003 BMD at
multiple sites and risk of fracture of multiple types: long-term results from the Study of Osteoporotic
Fractures. J Bone Miner Res 18:1947-1954.
46. Center JR, Bliuc D, Nguyen TV, Eisman JA 1/24/2007 Risk of subsequent fracture after low-trauma fracture in
men and women. JAMA 297:387-394.
47. Kanis JA, Johnell O, De Laet C, Johansson H, Oden A, Delmas P, Eisman J, Fujiwara S, Garnero P, Kroger H,
McCloskey EV, Mellstrom D, Melton LJ, Pols H, Reeve J, Silman A, Tenenhouse A 8/2004 A metaanalysis of previous fracture and subsequent fracture risk. Bone 35:375-382.
48. Huntjens KM, Kosar S, van Geel TA, Geusens PP, Willems P, Kessels A, Winkens B, Brink P, van HS 2/17/2010
Risk of subsequent fracture and mortality within 5 years after a non-vertebral fracture. Osteoporos Int.
49. van Geel TA, van HS, Geusens PP, Winkens B, Dinant GJ 1/2009 Clinical subsequent fractures cluster in time
after first fractures. Ann Rheum Dis 68:99-102.
50. Johnell O, Kanis JA, Oden A, Sernbo I, Redlund-Johnell I, Petterson C, De Laet C, Jonsson B 3/2004 Fracture
risk following an osteoporotic fracture. Osteoporos Int 15:175-179.
51. Lauritzen JB, Lund B 6/1993 Risk of hip fracture after osteoporosis fractures. 451 women with fracture of lumbar
spine, olecranon, knee or ankle. Acta Orthop Scand 64:297-300.
52. Mallmin H, Ljunghall S, Persson I, Naessen T, Krusemo UB, Bergstrom R 4/1993 Fracture of the distal forearm
as a forecaster of subsequent hip fracture: a population-based cohort study with 24 years of follow-up.
Calcif Tissue Int 52:269-272.
61
53. Ryg J, Rejnmark L, Overgaard S, Brixen K, Vestergaard P 7/2009 Hip fracture patients at risk of second hip
fracture: a nationwide population-based cohort study of 169,145 cases during 1977-2001. J Bone Miner
Res 24:1299-1307.
54. Court-Brown CM, Caesar B 8/2006 Epidemiology of adult fractures: A review. Injury 37:691-697.
55. Jonsson BY, Siggeirsdottir K, Mogensen B, Sigvaldason H, Sigursson G 4/2004 Fracture rate in a populationbased sample of men in Reykjavik. Acta Orthop Scand 75:195-200.
56. Maggi S, Kelsey JL, Litvak J, Heyse SP 9/1991 Incidence of hip fractures in the elderly: a cross-national analysis.
Osteoporos Int 1:232-241.
57. Kannus P, Parkkari J, Sievanen H, Heinonen A, Vuori I, Jarvinen M 1/1996 Epidemiology of hip fractures. Bone
18:57S-63S.
58. Xu L, Lu A, Zhao X, Chen X, Cummings SR 11/1/1996 Very low rates of hip fracture in Beijing, People's
Republic of China the Beijing Osteoporosis Project. Am J Epidemiol 144:901-907.
59. Nymark T, Lauritsen JM, Ovesen O, Rock ND, Jeune B 2/2006 Decreasing incidence of hip fracture in the Funen
County, Denmark. Acta Orthop 77:109-113.
60. Nevitt MC, Ettinger B, Black DM, Stone K, Jamal SA, Ensrud K, Segal M, Genant HK, Cummings SR 5/15/1998
The association of radiographically detected vertebral fractures with back pain and function: a prospective
study. Ann Intern Med 128:793-800.
61. Cooper C, Atkinson EJ, O'fallon WM, Melton LJ, III 2/1992 Incidence of clinically diagnosed vertebral fractures:
a population-based study in Rochester, Minnesota, 1985-1989. J Bone Miner Res 7:221-227.
62. Center JR, Nguyen TV, Schneider D, Sambrook PN, Eisman JA 3/13/1999 Mortality after all major types of
osteoporotic fracture in men and women: an observational study. Lancet 353:878-882.
63. Santavirta S, Konttinen YT, Heliovaara M, Knekt P, Luthje P, Aromaa A 4/1992 Determinants of osteoporotic
thoracic vertebral fracture. Screening of 57,000 Finnish women and men. Acta Orthop Scand 63:198-202.
64. O'Neill TW, Felsenberg D, Varlow J, Cooper C, Kanis JA, Silman AJ 7/1996 The prevalence of vertebral
deformity in european men and women: the European Vertebral Osteoporosis Study. J Bone Miner Res
11:1010-1018.
65. Samelson EJ, Hannan MT, Zhang Y, Genant HK, Felson DT, Kiel DP 8/2006 Incidence and risk factors for
vertebral fracture in women and men: 25-year follow-up results from the population-based Framingham
study. J Bone Miner Res 21:1207-1214.
66. Hasserius R, Karlsson MK, Nilsson BE, Redlund-Johnell I, Johnell O 1/2003 Prevalent vertebral deformities
predict increased mortality and increased fracture rate in both men and women: a 10-year populationbased study of 598 individuals from the Swedish cohort in the European Vertebral Osteoporosis Study.
Osteoporos Int 14:61-68.
67. Ioannidis G, Papaioannou A, Hopman WM, khtar-Danesh N, Anastassiades T, Pickard L, Kennedy CC, Prior JC,
Olszynski WP, Davison KS, Goltzman D, Thabane L, Gafni A, Papadimitropoulos EA, Brown JP, Josse
RG, Hanley DA, Adachi JD 9/1/2009 Relation between fractures and mortality: results from the
Canadian Multicentre Osteoporosis Study. CMAJ 181:265-271.
68. Bliuc D, Nguyen ND, Milch VE, Nguyen TV, Eisman JA, Center JR 2/4/2009 Mortality risk associated with lowtrauma osteoporotic fracture and subsequent fracture in men and women. JAMA 301:513-521.
69. Ismail AA, Pye SR, Cockerill WC, Lunt M, Silman AJ, Reeve J, Banzer D, Benevolenskaya LI, Bhalla A, Bruges AJ,
Cannata JB, Cooper C, Delmas PD, Dequeker J, Dilsen G, Falch JA, Felsch B, Felsenberg D, Finn JD,
Gennari C, Hoszowski K, Jajic I, Janott J, Johnell O, Kanis JA, Kragl G, Lopez VA, Lorenc R, Lyritis G,
62
Marchand F, Masaryk P, Matthis C, Miazgowski T, Naves-Diaz M, Pols HA, Poor G, Rapado A, Raspe
HH, Reid DM, Reisinger W, Scheidt-Nave C, Stepan J, Todd C, Weber K, Woolf AD, O'Neill TW
7/2002 Incidence of limb fracture across Europe: results from the European Prospective Osteoporosis
Study (EPOS). Osteoporos Int 13:565-571.
70. Lofthus CM, Frihagen F, Meyer HE, Nordsletten L, Melhuus K, Falch JA 6/2008 Epidemiology of distal forearm
fractures in Oslo, Norway. Osteoporos Int 19:781-786.
71. Haentjens P, Johnell O, Kanis JA, Bouillon R, Cooper C, Lamraski G, Vanderschueren D, Kaufman JM, Boonen S
12/2004 Evidence from data searches and life-table analyses for gender-related differences in absolute
risk of hip fracture after Colles' or spine fracture: Colles' fracture as an early and sensitive marker of
skeletal fragility in white men. J Bone Miner Res 19:1933-1944.
72. Ismail AA, Pye SR, Cockerill WC, Lunt M, Silman AJ, Reeve J, Banzer D, Benevolenskaya LI, Bhalla A, Bruges AJ,
Cannata JB, Cooper C, Delmas PD, Dequeker J, Dilsen G, Falch JA, Felsch B, Felsenberg D, Finn JD,
Gennari C, Hoszowski K, Jajic I, Janott J, Johnell O, Kanis JA, Kragl G, Lopez VA, Lorenc R, Lyritis G,
Marchand F, Masaryk P, Matthis C, Miazgowski T, Naves-Diaz M, Pols HA, Poor G, Rapado A, Raspe
HH, Reid DM, Reisinger W, Scheidt-Nave C, Stepan J, Todd C, Weber K, Woolf AD, O'Neill TW
7/2002 Incidence of limb fracture across Europe: results from the European Prospective Osteoporosis
Study (EPOS). Osteoporos Int 13:565-571.
73. Barrett-Connor E, Nielson CM, Orwoll E, Bauer DC, Cauley JA 2010 Epidemiology of rib fractures in older men:
Osteoporotic Fractures in Men (MrOS) prospective cohort study. BMJ 340:c1069.
74. Rapp K, Cameron ID, Kurrle S, Klenk J, Kleiner A, Heinrich S, Konig HH, Becker C 1/8/2010 Excess mortality
after pelvic fractures in institutionalized older people. Osteoporos Int.
75. Pocock NA, Eisman JA, Hopper JL, Yeates MG, Sambrook PN, Eberl S 9/1987 Genetic determinants of bone
mass in adults. A twin study. J Clin Invest 80:706-710.
76. Howard GM, Nguyen TV, Harris M, Kelly PJ, Eisman JA 8/1998 Genetic and environmental contributions to the
association between quantitative ultrasound and bone mineral density measurements: a twin study. J Bone
Miner Res 13:1318-1327.
77. Ralston SH, Galwey N, Mackay I, Albagha OM, Cardon L, Compston JE, Cooper C, Duncan E, Keen R, Langdahl
B, McLellan A, O'Riordan J, Pols HA, Reid DM, Uitterlinden AG, Wass J, Bennett ST 4/1/2005 Loci for
regulation of bone mineral density in men and women identified by genome wide linkage scan: the
FAMOS study. Hum Mol Genet 14:943-951.
78. Ioannidis JP, Ng MY, Sham PC, Zintzaras E, Lewis CM, Deng HW, Econs MJ, Karasik D, Devoto M, Kammerer
CM, Spector T, Andrew T, Cupples LA, Duncan EL, Foroud T, Kiel DP, Koller D, Langdahl B, Mitchell
BD, Peacock M, Recker R, Shen H, Sol-Church K, Spotila LD, Uitterlinden AG, Wilson SG, Kung AW,
Ralston SH 2/2007 Meta-analysis of genome-wide scans provides evidence for sex- and site-specific
regulation of bone mass. J Bone Miner Res 22:173-183.
79. Harvey NC, Javaid MK, Poole JR, Taylor P, Robinson SM, Inskip HM, Godfrey KM, Cooper C, Dennison EM
5/2008 Paternal skeletal size predicts intrauterine bone mineral accrual. J Clin Endocrinol Metab 93:16761681.
80. Ioannidis JP, Ralston SH, Bennett ST, Brandi ML, Grinberg D, Karassa FB, Langdahl B, van Meurs JB, Mosekilde
L, Scollen S, Albagha OM, Bustamante M, Carey AH, Dunning AM, Enjuanes A, van Leeuwen JP,
Mavilia C, Masi L, McGuigan FE, Nogues X, Pols HA, Reid DM, Schuit SC, Sherlock RE, Uitterlinden
AG 11/3/2004 Differential genetic effects of ESR1 gene polymorphisms on osteoporosis outcomes.
JAMA 292:2105-2114.
81. Langdahl BL, Uitterlinden AG, Ralston SH, Trikalinos TA, Balcells S, Brandi ML, Scollen S, Lips P, Lorenc R,
Obermayer-Pietsch B, Reid DM, Armas JB, Arp PP, Bassiti A, Bustamante M, Husted LB, Carey AH,
Perez CR, Dobnig H, Dunning AM, Fahrleitner-Pammer A, Falchetti A, Karczmarewicz E, Kruk M, van
63
Leeuwen JP, Masi L, van Meurs JB, Mangion J, McGuigan FE, Mellibovsky L, Mosekilde L, Nogues X,
Pols HA, Reeve J, Renner W, Rivadeneira F, van Schoor NM, Ioannidis JP 5/2008 Large-scale analysis of
association between polymorphisms in the transforming growth factor beta 1 gene (TGFB1) and
osteoporosis: the GENOMOS study. Bone 42:969-981.
82. van Meurs JB, Trikalinos TA, Ralston SH, Balcells S, Brandi ML, Brixen K, Kiel DP, Langdahl BL, Lips P,
Ljunggren O, Lorenc R, Obermayer-Pietsch B, Ohlsson C, Pettersson U, Reid DM, Rousseau F, Scollen
S, Van HW, Agueda L, Akesson K, Benevolenskaya LI, Ferrari SL, Hallmans G, Hofman A, Husted LB,
Kruk M, Kaptoge S, Karasik D, Karlsson MK, Lorentzon M, Masi L, McGuigan FE, Mellstrom D,
Mosekilde L, Nogues X, Pols HA, Reeve J, Renner W, Rivadeneira F, van Schoor NM, Weber K,
Ioannidis JP, Uitterlinden AG 3/19/2008 Large-scale analysis of association between LRP5 and LRP6
variants and osteoporosis. JAMA 299:1277-1290.
83. Ralston SH, Uitterlinden AG, Brandi ML, Balcells S, Langdahl BL, Lips P, Lorenc R, Obermayer-Pietsch B,
Scollen S, Bustamante M, Husted LB, Carey AH, ez-Perez A, Dunning AM, Falchetti A, Karczmarewicz
E, Kruk M, van Leeuwen JP, van Meurs JB, Mangion J, McGuigan FE, Mellibovsky L, Del MF, Pols HA,
Reeve J, Reid DM, Renner W, Rivadeneira F, van Schoor NM, Sherlock RE, Ioannidis JP 4/2006 Largescale evidence for the effect of the COLIA1 Sp1 polymorphism on osteoporosis outcomes: the
GENOMOS study. PLoS Med 3:e90.
84. Kiel DP, Demissie S, Dupuis J, Lunetta KL, Murabito JM, Karasik D 2007 Genome-wide association with bone
mass and geometry in the Framingham Heart Study. BMC Med Genet 8 Suppl 1:S14.
85. Styrkarsdottir U, Halldorsson BV, Gretarsdottir S, Gudbjartsson DF, Walters GB, Ingvarsson T, Jonsdottir T,
Saemundsdottir J, Center JR, Nguyen TV, Bagger Y, Gulcher JR, Eisman JA, Christiansen C, Sigurdsson
G, Kong A, Thorsteinsdottir U, Stefansson K 5/29/2008 Multiple genetic loci for bone mineral density
and fractures. N Engl J Med 358:2355-2365.
86. Ralston SH 3/2010 Genetics of osteoporosis. Ann N Y Acad Sci 1192:181-189.
87. Brixen K, Beckers S, Peeters A, Piters E, Balemans W, Nielsen TL, Wraae K, Bathum L, Brasen C, Hagen C,
Andersen M, Van HW, Abrahamsen B 12/2007 Polymorphisms in the Low-Density Lipoprotein
Receptor-Related Protein 5 (LRP5) Gene Are Associated with Peak Bone Mass in Non-sedentary Men:
Results from the Odense Androgen Study. Calcif Tissue Int 81:421-429.
88. Soroko SB, Barrett-Connor E, Edelstein SL, Kritz-Silverstein D 6/1994 Family history of osteoporosis and bone
mineral density at the axial skeleton: the Rancho Bernardo Study. J Bone Miner Res 9:761-769.
89. Cauley JA, Fullman RL, Stone KL, Zmuda JM, Bauer DC, Barrett-Connor E, Ensrud K, Lau EM, Orwoll ES
12/2005 Factors associated with the lumbar spine and proximal femur bone mineral density in older men.
Osteoporos Int 16:1525-1537.
90. Kanis JA, Johansson H, Oden A, Johnell O, De Laet C, Eisman JA, McCloskey EV, Mellstrom D, Melton LJ, III,
Pols HA, Reeve J, Silman AJ, Tenenhouse A 11/2004 A family history of fracture and fracture risk: a
meta-analysis. Bone 35:1029-1037.
91. Kanis JA, Johansson H, Oden A, Johnell O, De Laet C, Eisman JA, McCloskey EV, Mellstrom D, Melton LJ, III,
Pols HA, Reeve J, Silman AJ, Tenenhouse A 11/2004 A family history of fracture and fracture risk: a
meta-analysis. Bone 35:1029-1037.
92. Kelly PJ, Nguyen T, Hopper J, Pocock N, Sambrook P, Eisman J 1/1993 Changes in axial bone density with age: a
twin study. J Bone Miner Res 8:11-17.
93. Christian JC, Yu PL, Slemenda CW, Johnston CC, Jr. 3/1989 Heritability of bone mass: a longitudinal study in
aging male twins. Am J Hum Genet 44:429-433.
94. Michaelsson K, Melhus H, Ferm H, Ahlbom A, Pedersen NL 9/12/2005 Genetic liability to fractures in the
elderly. Arch Intern Med 165:1825-1830.
64
95. Bergvall N, Cnattingius S 9/2008 Familial (shared environmental and genetic) factors and the foetal origins of
cardiovascular diseases and type 2 diabetes: a review of the literature. J Intern Med 264:205-223.
96. Hattersley AT, Tooke JE 5/22/1999 The fetal insulin hypothesis: an alternative explanation of the association of
low birthweight with diabetes and vascular disease. Lancet 353:1789-1792.
97. Gale CR, Martyn CN, Kellingray S, Eastell R, Cooper C 1/2001 Intrauterine programming of adult body
composition. J Clin Endocrinol Metab 86:267-272.
98. Cooper C, Cawley M, Bhalla A, Egger P, Ring F, Morton L, Barker D 6/1995 Childhood growth, physical activity,
and peak bone mass in women. J Bone Miner Res 10:940-947.
99. Cooper C, Fall C, Egger P, Hobbs R, Eastell R, Barker D 1/1997 Growth in infancy and bone mass in later life.
Ann Rheum Dis 56:17-21.
100. Cooper C, Eriksson JG, Forsen T, Osmond C, Tuomilehto J, Barker DJ 2001 Maternal height, childhood growth
and risk of hip fracture in later life: a longitudinal study. Osteoporos Int 12:623-629.
101. Javaid MK, Crozier SR, Harvey NC, Gale CR, Dennison EM, Boucher BJ, Arden NK, Godfrey KM, Cooper C
1/7/2006 Maternal vitamin D status during pregnancy and childhood bone mass at age 9 years: a
longitudinal study. Lancet 367:36-43.
102. donald-Wallis C, Tobias JH, Davey SG, Lawlor DA 8/25/2010 Relation of maternal prepregnancy body mass
index with offspring bone mass in childhood: is there evidence for an intrauterine effect? Am J Clin Nutr.
103. Henry YM, Fatayerji D, Eastell R 4/2004 Attainment of peak bone mass at the lumbar spine, femoral neck and
radius in men and women: relative contributions of bone size and volumetric bone mineral density.
Osteoporos Int 15:263-273.
104. Eisman JA, Kelly PJ, Morrison NA, Pocock NA, Yeoman R, Birmingham J, Sambrook PN 1993 Peak bone mass
and osteoporosis prevention. Osteoporos Int 3 Suppl 1:56-60.
105. Bonjour JP, Theintz G, Buchs B, Slosman D, Rizzoli R 9/1991 Critical years and stages of puberty for spinal and
femoral bone mass accumulation during adolescence. J Clin Endocrinol Metab 73:555-563.
106. Nordstrom P, Neovius M, Nordstrom A 5/2007 Early and rapid bone mineral density loss of the proximal femur
in men. J Clin Endocrinol Metab 92:1902-1908.
107. Hoiberg M, Nielsen TL, Wraae K, Abrahamsen B, Hagen C, Andersen M, Brixen K 11/2007 Population-based
reference values for bone mineral density in young men. Osteoporos Int 18:1507-1514.
108. Kaptoge S, Da Silva JA, Brixen K, Reid DM, Kroger H, Nielsen TL, Andersen M, Hagen C, Lorenc R, Boonen S,
de Vernejoul MC, Stepan JJ, Adams J, Kaufman JM, Reeve J 8/2008 Geographical variation in DXA
bone mineral density in young European men and women. Results from the Network in Europe on Male
Osteoporosis (NEMO) study. Bone 43:332-339.
109. Henry YM, Fatayerji D, Eastell R 4/2004 Attainment of peak bone mass at the lumbar spine, femoral neck and
radius in men and women: relative contributions of bone size and volumetric bone mineral density.
Osteoporos Int 15:263-273.
110. Kirmani S, Christen D, van Lenthe GH, Fischer PR, Bouxsein ML, McCready LK, Melton LJ, III, Riggs BL, Amin
S, Muller R, Khosla S 6/2009 Bone structure at the distal radius during adolescent growth. J Bone Miner
Res 24:1033-1042.
111. Khosla S, Riggs BL, Atkinson EJ, Oberg AL, McDaniel LJ, Holets M, Peterson JM, Melton LJ, III 1/2006 Effects
of sex and age on bone microstructure at the ultradistal radius: a population-based noninvasive in vivo
assessment. J Bone Miner Res 21:124-131.
65
112. Cummings SR, Black DM, Nevitt MC, Browner W, Cauley J, Ensrud K, Genant HK, Palermo L, Scott J, Vogt TM
1/9/1993 Bone density at various sites for prediction of hip fractures. The Study of Osteoporotic
Fractures Research Group. Lancet 341:72-75.
113. Marshall D, Johnell O, Wedel H 5/18/1996 Meta-analysis of how well measures of bone mineral density predict
occurrence of osteoporotic fractures. BMJ 312:1254-1259.
114. Hopper JL, Green RM, Nowson CA, Young D, Sherwin AJ, Kaymakci B, Larkins RG, Wark JD 1/1/1998
Genetic, common environment, and individual specific components of variance for bone mineral density
in 10- to 26-year-old females: a twin study. Am J Epidemiol 147:17-29.
115. Hind K, Burrows M 1/2007 Weight-bearing exercise and bone mineral accrual in children and adolescents: a
review of controlled trials. Bone 40:14-27.
116. Winzenberg T, Shaw K, Fryer J, Jones G 10/14/2006 Effects of calcium supplementation on bone density in
healthy children: meta-analysis of randomised controlled trials. BMJ 333:775.
117. Kremer R, Campbell PP, Reinhardt T, Gilsanz V 1/2009 Vitamin D status and its relationship to body fat, final
height, and peak bone mass in young women. J Clin Endocrinol Metab 94:67-73.
118. Nguyen TV, Eisman JA, Kelly PJ, Sambrook PN 8/1/1996 Risk factors for osteoporotic fractures in elderly men.
Am J Epidemiol 144:255-263.
119. Hannan MT, Felson DT, wson-Hughes B, Tucker KL, Cupples LA, Wilson PW, Kiel DP 4/2000 Risk factors for
longitudinal bone loss in elderly men and women: the Framingham Osteoporosis Study. J Bone Miner
Res 15:710-720.
120. Burger H, De Laet CE, van Daele PL, Weel AE, Witteman JC, Hofman A, Pols HA 5/1/1998 Risk factors for
increased bone loss in an elderly population: the Rotterdam Study. Am J Epidemiol 147:871-879.
121. Dennison E, Eastell R, Fall CH, Kellingray S, Wood PJ, Cooper C 1999 Determinants of bone loss in elderly men
and women: a prospective population-based study. Osteoporos Int 10:384-391.
122. Szulc P, Marchand F, Duboeuf F, Delmas PD 2/2000 Cross-sectional assessment of age-related bone loss in men:
the MINOS study. Bone 26:123-129.
123. Cawthon PM, Ewing SK, McCulloch CE, Ensrud KE, Cauley JA, Cummings SR, Orwoll ES 10/2009 Loss of hip
BMD in older men: the osteoporotic fractures in men (MrOS) study. J Bone Miner Res 24:1728-1735.
124. Kaptoge S, Reid DM, Scheidt-Nave C, Poor G, Pols HA, Khaw KT, Felsenberg D, Benevolenskaya LI, Diaz MN,
Stepan JJ, Eastell R, Boonen S, Cannata JB, Glueer CC, Crabtree NJ, Kaufman JM, Reeve J 3/2007
Geographic and other determinants of BMD change in European men and women at the hip and spine. a
population-based study from the Network in Europe for Male Osteoporosis (NEMO)
7. Bone 40:662-673.
125. Yoshimura N, Kinoshita H, Danjoh S, Takijiri T, Morioka S, Kasamatsu T, Sakata K, Hashimoto T 10/2002 Bone
loss at the lumbar spine and the proximal femur in a rural Japanese community, 1990-2000: the Miyama
study. Osteoporos Int 13:803-808.
126. Riggs BL, Melton IL, III, Robb RA, Camp JJ, Atkinson EJ, Peterson JM, Rouleau PA, McCollough CH, Bouxsein
ML, Khosla S 12/2004 Population-based study of age and sex differences in bone volumetric density,
size, geometry, and structure at different skeletal sites. J Bone Miner Res 19:1945-1954.
127. Macdonald HM, Nishiyama KK, Kang J, Hanley DA, Boyd SK 6/30/2010 Age-related patterns of trabecular and
cortical bone loss differ between sexes and skeletal sites: A population-based HR-pQCT study. J Bone
Miner Res.
66
128. Riggs BL, Melton LJ, Robb RA, Camp JJ, Atkinson EJ, McDaniel L, Amin S, Rouleau PA, Khosla S 2/2008 A
population-based assessment of rates of bone loss at multiple skeletal sites: evidence for substantial
trabecular bone loss in young adult women and men. J Bone Miner Res 23:205-214.
129. Marshall LM, Lang TF, Lambert LC, Zmuda JM, Ensrud KE, Orwoll ES 8/2006 Dimensions and volumetric
BMD of the proximal femur and their relation to age among older U.S. men. J Bone Miner Res 21:11971206.
130. Duan Y, Turner CH, Kim BT, Seeman E 12/2001 Sexual dimorphism in vertebral fragility is more the result of
gender differences in age-related bone gain than bone loss. J Bone Miner Res 16:2267-2275.
131. Keaveny TM, Kopperdahl D, Melton L, Hoffmann P, Amin S, Riggs B, Khosla S 10/29/2009 Age-Dependence of
Femoral Strength in White Women and Men. J Bone Miner Res.
132. Bouxsein ML, Melton LJ, III, Riggs BL, Muller J, Atkinson EJ, Oberg AL, Robb RA, Camp JJ, Rouleau PA,
McCollough CH, Khosla S 9/2006 Age- and sex-specific differences in the factor of risk for vertebral
fracture: a population-based study using QCT. J Bone Miner Res 21:1475-1482.
133. Burghardt AJ, Kazakia GJ, Ramachandran S, Link TM, Majumdar S 5/2010 Age- and gender-related differences in
the geometric properties and biomechanical significance of intracortical porosity in the distal radius and
tibia. J Bone Miner Res 25:983-993.
134. Silva MJ, Gibson LJ 8/1997 Modeling the mechanical behavior of vertebral trabecular bone: effects of age-related
changes in microstructure. Bone 21:191-199.
135. Riggs BL, Khosla S, Melton LJ, III 6/2002 Sex steroids and the construction and conservation of the adult
skeleton. Endocr Rev 23:279-302.
136. Greendale GA, Edelstein S, Barrett-Connor E 11/1997 Endogenous sex steroids and bone mineral density in older
women and men: the Rancho Bernardo Study. J Bone Miner Res 12:1833-1843.
137. Orwoll E, Lambert LC, Marshall LM, Phipps K, Blank J, Barrett-Connor E, Cauley J, Ensrud K, Cummings S
4/2006 Testosterone and estradiol among older men. J Clin Endocrinol Metab 91:1336-1344.
138. Leder BZ, LeBlanc KM, Schoenfeld DA, Eastell R, Finkelstein JS 1/2003 Differential effects of androgens and
estrogens on bone turnover in normal men. J Clin Endocrinol Metab 88:204-210.
139. Falahati-Nini A, Riggs BL, Atkinson EJ, O'fallon WM, Eastell R, Khosla S 12/2000 Relative contributions of
testosterone and estrogen in regulating bone resorption and formation in normal elderly men. J Clin
Invest 106:1553-1560.
140. Slemenda CW, Longcope C, Zhou L, Hui SL, Peacock M, Johnston CC 10/1/1997 Sex steroids and bone mass in
older men. Positive associations with serum estrogens and negative associations with androgens. J Clin
Invest 100:1755-1759.
141. Vanderschueren D, Pye SR, Venken K, Borghs H, Gaytant J, Huhtaniemi IT, Adams JE, Ward KA, Bartfai G,
Casanueva FF, Finn JD, Forti G, Giwercman A, Han TS, Kula K, Labrie F, Lean ME, Pendleton N,
Punab M, Silman AJ, Wu FC, O'Neill TW, Boonen S 12/15/2009 Gonadal sex steroid status and bone
health in middle-aged and elderly European men. Osteoporos Int.
142. Szulc P, Munoz F, Claustrat B, Garnero P, Marchand F, Duboeuf F, Delmas PD 1/2001 Bioavailable estradiol
may be an important determinant of osteoporosis in men: the MINOS study. J Clin Endocrinol Metab
86:192-199.
143. Amin S, Zhang Y, Sawin CT, Evans SR, Hannan MT, Kiel DP, Wilson PW, Felson DT 12/19/2000 Association
of hypogonadism and estradiol levels with bone mineral density in elderly men from the Framingham
study. Ann Intern Med 133:951-963.
67
144. Amin S, Zhang Y, Felson DT, Sawin CT, Hannan MT, Wilson PW, Kiel DP 5/2006 Estradiol, testosterone, and
the risk for hip fractures in elderly men from the Framingham Study. Am J Med 119:426-433.
145. Fink HA, Ewing SK, Ensrud KE, Barrett-Connor E, Taylor BC, Cauley JA, Orwoll ES 10/2006 Association of
testosterone and estradiol deficiency with osteoporosis and rapid bone loss in older men. J Clin
Endocrinol Metab 91:3908-3915.
146. Leblanc ES, Nielson CM, Marshall LM, Lapidus JA, Barrett-Connor E, Ensrud KE, Hoffman AR, Laughlin G,
Ohlsson C, Orwoll ES 9/2009 The effects of serum testosterone, estradiol, and sex hormone binding
globulin levels on fracture risk in older men. J Clin Endocrinol Metab 94:3337-3346.
147. Meier C, Nguyen TV, Handelsman DJ, Schindler C, Kushnir MM, Rockwood AL, Meikle AW, Center JR, Eisman
JA, Seibel MJ 1/14/2008 Endogenous sex hormones and incident fracture risk in older men: the Dubbo
Osteoporosis Epidemiology Study. Arch Intern Med 168:47-54.
148. Bjornerem A, Ahmed LA, Joakimsen RM, Berntsen GK, Fonnebo V, Jorgensen L, Oian P, Seeman E, Straume B
7/2007 A prospective study of sex steroids, sex hormone-binding globulin, and non-vertebral fractures in
women and men: the Tromso Study. Eur J Endocrinol 157:119-125.
149. Khosla S, Melton LJ, III, Atkinson EJ, O'fallon WM 8/2001 Relationship of serum sex steroid levels to
longitudinal changes in bone density in young versus elderly men. J Clin Endocrinol Metab 86:3555-3561.
150. Smith MR 9/2002 Osteoporosis during androgen deprivation therapy for prostate cancer. Urology 60:79-85.
151. Shahinian VB, Kuo YF, Freeman JL, Goodwin JS 1/13/2005 Risk of fracture after androgen deprivation for
prostate cancer. N Engl J Med 352:154-164.
152. Abrahamsen B, Nielsen MF, Eskildsen P, Andersen JT, Walter S, Brixen K 10/2007 Fracture risk in Danish men
with prostate cancer: a nationwide register study. BJU Int 100:749-754.
153. Papaioannou A, Kennedy CC, Cranney A, Hawker G, Brown JP, Kaiser SM, Leslie WD, O'Brien CJ, Sawka AM,
Khan A, Siminoski K, Tarulli G, Webster D, McGowan J, Adachi JD 4/2009 Risk factors for low BMD
in healthy men age 50 years or older: a systematic review. Osteoporos Int 20:507-518.
154. De Laet C, Kanis JA, Oden A, Johanson H, Johnell O, Delmas P, Eisman JA, Kroger H, Fujiwara S, Garnero P,
McCloskey EV, Mellstrom D, Melton LJ, III, Meunier PJ, Pols HA, Reeve J, Silman A, Tenenhouse A
11/2005 Body mass index as a predictor of fracture risk: a meta-analysis. Osteoporos Int 16:1330-1338.
155. De Laet C, Kanis JA, Oden A, Johanson H, Johnell O, Delmas P, Eisman JA, Kroger H, Fujiwara S, Garnero P,
McCloskey EV, Mellstrom D, Melton LJ, III, Meunier PJ, Pols HA, Reeve J, Silman A, Tenenhouse A
11/2005 Body mass index as a predictor of fracture risk: a meta-analysis. Osteoporos Int 16:1330-1338.
156. Wilsgaard T, Emaus N, Ahmed LA, Grimnes G, Joakimsen RM, Omsland TK, Berntsen GR 4/1/2009 Lifestyle
impact on lifetime bone loss in women and men: the Tromso Study. Am J Epidemiol 169:877-886.
157. Ensrud KE, Fullman RL, Barrett-Connor E, Cauley JA, Stefanick ML, Fink HA, Lewis CE, Orwoll E 4/2005
Voluntary weight reduction in older men increases hip bone loss: the osteoporotic fractures in men study.
J Clin Endocrinol Metab 90:1998-2004.
158. Wilsgaard T, Jacobsen BK, Ahmed LA, Joakimsen RM, Stormer J, Jorgensen L 6/12/2010 BMI change is
associated with fracture incidence, but only in non-smokers. The Tromso Study. Osteoporos Int.
159. Nielson CM, Marshall LM, Adams AL, Leblanc ES, Cawthon PM, Ensrud K, Stefanick ML, Barrett-Connor E,
Orwoll ES 9/2/2010 BMI and fracture risk in older men: the osteoporotic fractures in men (MrOS)
study. J Bone Miner Res.
160. Hippisley-Cox J, Coupland C 2009 Predicting risk of osteoporotic fracture in men and women in England and
Wales: prospective derivation and validation of QFractureScores. BMJ 339:b4229.
68
161. Holmberg AH, Johnell O, Nilsson PM, Nilsson J, Berglund G, Akesson K 2006 Risk factors for fragility fracture
in middle age. A prospective population-based study of 33,000 men and women. Osteoporos Int 17:10651077.
162. Orwoll E, Nielson CM, Marshall LM, Lambert L, Holton KF, Hoffman AR, Barrett-Connor E, Shikany JM, Dam
T, Cauley JA 4/2009 Vitamin D deficiency in older men. J Clin Endocrinol Metab 94:1214-1222.
163. Friedmann JM, Elasy T, Jensen GL 4/2001 The relationship between body mass index and self-reported functional
limitation among older adults: a gender difference. J Am Geriatr Soc 49:398-403.
164. Travison TG, Araujo AB, Esche GR, Beck TJ, McKinlay JB 2/2008 Lean mass and not fat mass is associated with
male proximal femur strength. J Bone Miner Res 23:189-198.
165. Tervo T, Nordstrom P, Neovius M, Nordstrom A 12/2008 Constant adaptation of bone to current physical
activity level in men: a 12-year longitudinal study. J Clin Endocrinol Metab 93:4873-4879.
166. Nordstrom A, Hogstrom M, Nordstrom P 3/2008 Effects of different types of weight-bearing loading on bone
mass and size in young males: a longitudinal study. Bone 42:565-571.
167. Tervo T, Nordstrom P, Neovius M, Nordstrom A 12/2009 Reduced physical activity corresponds with greater
bone loss at the trabecular than the cortical bone sites in men. Bone 45:1073-1078.
168. Nordstrom A, Olsson T, Nordstrom P 7/2006 Sustained benefits from previous physical activity on bone mineral
density in males. J Clin Endocrinol Metab 91:2600-2604.
169. Daly RM, Ahlborg HG, Ringsberg K, Gardsell P, Sernbo I, Karlsson MK 12/2008 Association between changes in
habitual physical activity and changes in bone density, muscle strength, and functional performance in
elderly men and women. J Am Geriatr Soc 56:2252-2260.
170. Morseth B, Emaus N, Wilsgaard T, Jacobsen BK, Jorgensen L 5/2010 Leisure time physical activity in adulthood
is positively associated with bone mineral density 22 years later. The Tromso study. Eur J Epidemiol
25:325-331.
171. Paganini-Hill A, Chao A, Ross RK, Henderson BE 1/1991 Exercise and other factors in the prevention of hip
fracture: the Leisure World study. Epidemiology 2:16-25.
172. Kujala UM, Kaprio J, Kannus P, Sarna S, Koskenvuo M 3/13/2000 Physical activity and osteoporotic hip fracture
risk in men. Arch Intern Med 160:705-708.
173. Silman AJ, O'Neill TW, Cooper C, Kanis J, Felsenberg D 5/1997 Influence of physical activity on vertebral
deformity in men and women: results from the European Vertebral Osteoporosis Study. J Bone Miner
Res 12:813-819.
174. Lips P 8/2001 Vitamin D deficiency and secondary hyperparathyroidism in the elderly: consequences for bone loss
and fractures and therapeutic implications
115. Endocr Rev 22:477-501.
175. Guillemant J, Le HT, Maria A, Allemandou A, Peres G, Guillemant S 2001 Wintertime vitamin D deficiency in
male adolescents: effect on parathyroid function and response to vitamin D3 supplements. Osteoporos
Int 12:875-879.
176. Valimaki VV, Alfthan H, Lehmuskallio E, Loyttyniemi E, Sahi T, Stenman UH, Suominen H, Valimaki MJ 1/2004
Vitamin D status as a determinant of peak bone mass in young Finnish men. J Clin Endocrinol Metab
89:76-80.
177. Norman AW, Bouillon R, Whiting SJ, Vieth R, Lips P 3/2007 13th Workshop consensus for vitamin D nutritional
guidelines. J Steroid Biochem Mol Biol 103:204-205.
69
178. Melhus H, Snellman G, Gedeborg R, Byberg L, Berglund L, Mallmin H, Hellman P, Blomhoff R, Hagstrom E,
Arnlov J, Michaelsson K 6/2010 Plasma 25-hydroxyvitamin D levels and fracture risk in a communitybased cohort of elderly men in Sweden. J Clin Endocrinol Metab 95:2637-2645.
179. 2010 Patient level pooled analysis of 68 500 patients from seven major vitamin D fracture trials in US and Europe.
BMJ 340:b5463.
180. Bischoff-Ferrari HA, Willett WC, Wong JB, Stuck AE, Staehelin HB, Orav EJ, Thoma A, Kiel DP, Henschkowski
J 3/23/2009 Prevention of nonvertebral fractures with oral vitamin D and dose dependency: a metaanalysis of randomized controlled trials. Arch Intern Med 169:551-561.
181. Saquib N, von MD, Garland CF, Barrett-Connor E 12/2006 Serum 25-hydroxyvitamin D, parathyroid hormone,
and bone mineral density in men: the Rancho Bernardo study. Osteoporos Int 17:1734-1741.
182. Ensrud KE, Taylor BC, Paudel ML, Cauley JA, Cawthon PM, Cummings SR, Fink HA, Barrett-Connor E, Zmuda
JM, Shikany JM, Orwoll ES 8/2009 Serum 25-hydroxyvitamin D levels and rate of hip bone loss in older
men. J Clin Endocrinol Metab 94:2773-2780.
183. Cauley JA, Parimi N, Ensrud KE, Bauer DC, Cawthon PM, Cummings SR, Hoffman AR, Shikany JM, BarrettConnor E, Orwoll E 9/23/2009 Serum 25 HydroxyVitamin D and the Risk of Hip and Non-spine
Fractures in Older Men. J Bone Miner Res.
184. Bischoff-Ferrari HA, Dietrich T, Orav EJ, wson-Hughes B 5/1/2004 Positive association between 25-hydroxy
vitamin D levels and bone mineral density: a population-based study of younger and older adults. Am J
Med 116:634-639.
185. Hannan MT, Litman HJ, Araujo AB, McLennan CE, McLean RR, McKinlay JB, Chen TC, Holick MF 11/6/2007
Serum 25-Hydroxyvitamin D and Bone Mineral Density in a Racially and Ethnically Diverse Group of
Men. J Clin Endocrinol Metab.
186. Muscogiuri G, Sorice GP, Prioletta A, Policola C, Casa SD, Pontecorvi A, Giaccari A 2/11/2010 25Hydroxyvitamin D Concentration Correlates With Insulin-Sensitivity and BMI in Obesity. Obesity (Silver
Spring).
187. Lenders CM, Feldman HA, Von SE, Merewood A, Sweeney C, Wilson DM, Lee PD, Abrams SH, Gitelman SE,
Wertz MS, Klish WJ, Taylor GA, Chen TC, Holick MF 9/2009 Relation of body fat indexes to vitamin D
status and deficiency among obese adolescents. Am J Clin Nutr 90:459-467.
188. Lee DM, Rutter MK, O'Neill TW, Boonen S, Vanderschueren D, Bouillon R, Bartfai G, Casanueva FF, Finn JD,
Forti G, Giwercman A, Han TS, Huhtaniemi IT, Kula K, Lean ME, Pendleton N, Punab M, Silman AJ,
Wu FC 12/2009 Vitamin D, parathyroid hormone and the metabolic syndrome in middle-aged and older
European men. Eur J Endocrinol 161:947-954.
189. Cheng S, Massaro JM, Fox CS, Larson MG, Keyes MJ, McCabe EL, Robins SJ, O'donnell CJ, Hoffmann U,
Jacques PF, Booth SL, Vasan RS, Wolf M, Wang TJ 1/2010 Adiposity, cardiometabolic risk, and vitamin
D status: the Framingham Heart Study. Diabetes 59:242-248.
190. Brot C, Vestergaard P, Kolthoff N, Gram J, Hermann AP, Sorensen OH 8/2001 Vitamin D status and its
adequacy in healthy Danish perimenopausal women: relationships to dietary intake, sun exposure and
serum parathyroid hormone. Br J Nutr 86 Suppl 1:S97-103.
191. Bischoff HA, Stahelin HB, Dick W, Akos R, Knecht M, Salis C, Nebiker M, Theiler R, Pfeifer M, Begerow B, Lew
RA, Conzelmann M 2/2003 Effects of vitamin D and calcium supplementation on falls: a randomized
controlled trial. J Bone Miner Res 18:343-351.
192. Bergstrom U, Bjornstig U, Stenlund H, Jonsson H, Svensson O 9/2008 Fracture mechanisms and fracture pattern
in men and women aged 50 years and older: a study of a 12-year population-based injury register, Umea,
Sweden. Osteoporos Int 19:1267-1273.
70
193. Sattin RW, Lambert Huber DA, Devito CA, Rodriguez JG, Ros A, Bacchelli S, Stevens JA, Waxweiler RJ 6/1990
The incidence of fall injury events among the elderly in a defined population. Am J Epidemiol 131:10281037.
194. Kannus P, Sievanen H, Palvanen M, Jarvinen T, Parkkari J 11/26/2005 Prevention of falls and consequent injuries
in elderly people. Lancet 366:1885-1893.
195. Gates S, Fisher JD, Cooke MW, Carter YH, Lamb SE 1/19/2008 Multifactorial assessment and targeted
intervention for preventing falls and injuries among older people in community and emergency care
settings: systematic review and meta-analysis. BMJ 336:130-133.
196. Gillespie LD, Robertson MC, Gillespie WJ, Lamb SE, Gates S, Cumming RG, Rowe BH 2009 Interventions for
preventing falls in older people living in the community. Cochrane Database Syst Rev:CD007146.
197. Cameron ID, Murray GR, Gillespie LD, Robertson MC, Hill KD, Cumming RG, Kerse N 2010 Interventions for
preventing falls in older people in nursing care facilities and hospitals. Cochrane Database Syst
Rev:CD005465.
198. Pfeifer M, Begerow B, Minne HW, Suppan K, Fahrleitner-Pammer A, Dobnig H 2/2009 Effects of a long-term
vitamin D and calcium supplementation on falls and parameters of muscle function in communitydwelling older individuals. Osteoporos Int 20:315-322.
199. Bischoff-Ferrari HA, wson-Hughes B, Staehelin HB, Orav JE, Stuck AE, Theiler R, Wong JB, Egli A, Kiel DP,
Henschkowski J 2009 Fall prevention with supplemental and active forms of vitamin D: a meta-analysis
of randomised controlled trials. BMJ 339:b3692.
200. Kanis JA, Johnell O, Oden A, Johansson H, De Laet C, Eisman JA, Fujiwara S, Kroger H, McCloskey EV,
Mellstrom D, Melton LJ, Pols H, Reeve J, Silman A, Tenenhouse A 2/2005 Smoking and fracture risk: a
meta-analysis. Osteoporos Int 16:155-162.
201. Szulc P, Garnero P, Claustrat B, Marchand F, Duboeuf F, Delmas PD 2/2002 Increased bone resorption in
moderate smokers with low body weight: the Minos study. J Clin Endocrinol Metab 87:666-674.
202. Liu H, Paige NM, Goldzweig CL, Wong E, Zhou A, Suttorp MJ, Munjas B, Orwoll E, Shekelle P 5/6/2008
Screening for osteoporosis in men: a systematic review for an American College of Physicians guideline.
Ann Intern Med 148:685-701.
203. Roy DK, O'neill TW, Finn JD, Lunt M, Silman AJ, Felsenberg D, Armbrecht G, Banzer D, Benevolenskaya LI,
Bhalla A, Bruges AJ, Cannata JB, Cooper C, Dequeker J, Diaz MN, Eastell R, Yershova OB, Felsch B,
Gowin W, Havelka S, Hoszowski K, Ismail AA, Jajic I, Janott I, Johnell O, Kanis JA, Kragl G, Lopez
VA, Lorenc R, Lyritis G, Masaryk P, Matthis C, Miazgowski T, Gennari C, Pols HA, Poor G, Raspe HH,
Reid DM, Reisinger W, Scheidt-Nave C, Stepan JJ, Todd CJ, Weber K, Woolf AD, Reeve J 1/2003
Determinants of incident vertebral fracture in men and women: results from the European Prospective
Osteoporosis Study (EPOS). Osteoporos Int 14:19-26.
204. Kanis JA, Johansson H, Johnell O, Oden A, De Laet C, Eisman JA, Pols H, Tenenhouse A 7/2005 Alcohol intake
as a risk factor for fracture. Osteoporos Int 16:737-742.
205. Hoidrup S, Gronbaek M, Gottschau A, Lauritzen JB, Schroll M 6/1/1999 Alcohol intake, beverage preference,
and risk of hip fracture in men and women. Copenhagen Centre for Prospective Population Studies. Am
J Epidemiol 149:993-1001.
206. Vestergaard P 4/2007 Discrepancies in bone mineral density and fracture risk in patients with type 1 and type 2
diabetes--a meta-analysis. Osteoporos Int 18:427-444.
207. Petit MA, Paudel ML, Taylor BC, Hughes JM, Strotmeyer ES, Schwartz AV, Cauley JA, Zmuda JM, Hoffman AR,
Ensrud KE 7/13/2009 Bone Mass and Strength in Older Men with Type 2 Diabetes: The Osteoporotic
Fractures in Men Study. J Bone Miner Res.
71
208. Schwartz AV, Vittinghoff E, Sellmeyer DE, Feingold KR, de RN, Strotmeyer ES, Shorr RI, Vinik AI, Odden MC,
Park SW, Faulkner KA, Harris TB 3/2008 Diabetes-related complications, glycemic control, and falls in
older adults. Diabetes Care 31:391-396.
209. Huang ES, Karter AJ, Danielson KK, Warton EM, Ahmed AT 2/2010 The association between the number of
prescription medications and incident falls in a multi-ethnic population of adult type-2 diabetes patients:
the diabetes and aging study. J Gen Intern Med 25:141-146.
210. Sennerby U, Melhus H, Gedeborg R, Byberg L, Garmo H, Ahlbom A, Pedersen NL, Michaelsson K 10/21/2009
Cardiovascular diseases and risk of hip fracture. JAMA 302:1666-1673.
211. Szulc P, Samelson EJ, Kiel DP, Delmas PD 12/2009 Increased bone resorption is associated with increased risk of
cardiovascular events in men: the MINOS study. J Bone Miner Res 24:2023-2031.
212. Szulc P, Kiel DP, Delmas PD 1/2008 Calcifications in the abdominal aorta predict fractures in men: MINOS
study. J Bone Miner Res 23:95-102.
213. Vestergaard P, Rejnmark L, Mosekilde L 2/2009 Hypertension is a risk factor for fractures. Calcif Tissue Int
84:103-111.
214. Dam TT, Harrison S, Fink HA, Ramsdell J, Barrett-Connor E 10/9/2009 Bone mineral density and fractures in
older men with chronic obstructive pulmonary disease or asthma. Osteoporos Int.
215. Guler-Yuksel M, Allaart CF, Goekoop-Ruiterman YP, de Vries-Bouwstra JK, van Groenendael JH, Mallee C, de
Bois MH, Breedveld FC, Dijkmans BA, Lems WF 3/2009 Changes in hand and generalised bone mineral
density in patients with recent-onset rheumatoid arthritis. Ann Rheum Dis 68:330-336.
216. Furuya T, Kotake S, Inoue E, Nanke Y, Yago T, Hara M, Tomatsu T, Kamatani N, Yamanaka H 2008 Risk factors
associated with incident fractures in Japanese men with rheumatoid arthritis: a prospective observational
cohort study. J Bone Miner Metab 26:499-505.
217. Abrahamsen B, Brixen K 4/2009 Mapping the prescriptiome to fractures in men--a national analysis of
prescription history and fracture risk. Osteoporos Int 20:585-597.
218. de VF, Pouwels S, Lammers JW, Leufkens HG, Bracke M, Cooper C, van Staa TP 2/2007 Use of inhaled and oral
glucocorticoids, severity of inflammatory disease and risk of hip/femur fracture: a population-based casecontrol study. J Intern Med 261:170-177.
219. Drummond MB, Dasenbrook EC, Pitz MW, Murphy DJ, Fan E 11/26/2008 Inhaled corticosteroids in patients
with stable chronic obstructive pulmonary disease: a systematic review and meta-analysis. JAMA
300:2407-2416.
220. Vestergaard P, Rejnmark L, Mosekilde L 4/2005 Fracture risk associated with systemic and topical corticosteroids.
J Intern Med 257:374-384.
221. Richards JB, Papaioannou A, Adachi JD, Joseph L, Whitson HE, Prior JC, Goltzman D 1/22/2007 Effect of
selective serotonin reuptake inhibitors on the risk of fracture. Arch Intern Med 167:188-194.
222. Haney EM, Chan BK, Diem SJ, Ensrud KE, Cauley JA, Barrett-Connor E, Orwoll E, Bliziotes MM 6/25/2007
Association of low bone mineral density with selective serotonin reuptake inhibitor use by older men.
Arch Intern Med 167:1246-1251.
223. Yadav VK, Oury F, Suda N, Liu ZW, Gao XB, Confavreux C, Klemenhagen KC, Tanaka KF, Gingrich JA, Guo
XE, Tecott LH, Mann JJ, Hen R, Horvath TL, Karsenty G 9/4/2009 A serotonin-dependent mechanism
explains the leptin regulation of bone mass, appetite, and energy expenditure. Cell 138:976-989.
224. Vestergaard P, Tigaran S, Rejnmark L, Tigaran C, Dam M, Mosekilde L 5/1999 Fracture risk is increased in
epilepsy. Acta Neurol Scand 99:269-275.
72
225. Andress DL, Ozuna J, Tirschwell D, Grande L, Johnson M, Jacobson AF, Spain W 5/2002 Antiepileptic druginduced bone loss in young male patients who have seizures. Arch Neurol 59:781-786.
226. Ensrud KE, Walczak TS, Blackwell TL, Ensrud ER, Barrett-Connor E, Orwoll ES 9/2/2008 Antiepileptic drug
use and rates of hip bone loss in older men: a prospective study. Neurology 71:723-730.
227. Danish Bone Society 2009 http://www.dkms.dk/PDF/DKMS_Osteoporose_2009.pdf.
228. Rubin KH, Abrahamsen B, Hermann AP, Bech M, Gram J, Brixen K 8/4/2010 Prevalence of risk factors for
fractures and use of DXA scanning in Danish women. A regional population-based study. Osteoporos
Int.
229. Vestergaard P, Rejnmark L, Mosekilde L 2006 Anxiolytics, sedatives, antidepressants, neuroleptics and the risk of
fracture. Osteoporos Int 17:807-816.
230. McCloskey EV, Johansson H, Oden A, Vasireddy S, Kayan K, Pande K, Jalava T, Kanis JA 5/2009 Ten-year
fracture probability identifies women who will benefit from clodronate therapy--additional results from a
double-blind, placebo-controlled randomised study. Osteoporos Int 20:811-817.
231. Kanis JA, Johansson H, Oden A, McCloskey EV 6/2009 Bazedoxifene reduces vertebral and clinical fractures in
postmenopausal women at high risk assessed with FRAX. Bone 44:1049-1054.
232. Sandhu SK, Nguyen ND, Center JR, Pocock NA, Eisman JA, Nguyen TV 7/25/2009 Prognosis of fracture:
evaluation of predictive accuracy of the FRAX algorithm and Garvan nomogram. Osteoporos Int.
233. Schurch MA, Rizzoli R, Mermillod B, Vasey H, Michel JP, Bonjour JP 12/1996 A prospective study on
socioeconomic aspects of fracture of the proximal femur. J Bone Miner Res 11:1935-1942.
234. Chrischilles EA, Butler CD, Davis CS, Wallace RB 10/1991 A model of lifetime osteoporosis impact. Arch Intern
Med 151:2026-2032.
235. Poor G, Atkinson EJ, O'fallon WM, Melton LJ, III 10/1995 Determinants of reduced survival following hip
fractures in men. Clin Orthop Relat Res:260-265.
236. Kannegaard PN, van der MS, Eiken P, Abrahamsen B 3/2010 Excess mortality in men compared with women
following a hip fracture. National analysis of comedications, comorbidity and survival. Age Ageing
39:203-209.
237. Forsen L, Sogaard AJ, Meyer HE, Edna T, Kopjar B 1999 Survival after hip fracture: short- and long-term excess
mortality according to age and gender. Osteoporos Int 10:73-78.
238. Bass E, French DD, Bradham DD, Rubenstein LZ 7/2007 Risk-adjusted mortality rates of elderly veterans with
hip fractures. Ann Epidemiol 17:514-519.
239. Haentjens P, Magaziner J, Colon-Emeric CS, Vanderschueren D, Milisen K, Velkeniers B, Boonen S 3/16/2010
Meta-analysis: excess mortality after hip fracture among older women and men. Ann Intern Med 152:380390.
240. Myers AH, Robinson EG, Van Natta ML, Michelson JD, Collins K, Baker SP 11/15/1991 Hip fractures among
the elderly: factors associated with in-hospital mortality. Am J Epidemiol 134:1128-1137.
241. Ewald DP, Eisman JA, Ewald BD, Winzenberg TM, Seibel MJ, Ebeling PR, Flicker LA, Nash PT 2/2/2009
Population rates of bone densitometry use in Australia, 2001-2005, by sex and rural versus urban location.
Med J Aust 190:126-128.
242. Jones G, Nguyen T, Sambrook PN, Kelly PJ, Gilbert C, Eisman JA 9/1994 Symptomatic fracture incidence in
elderly men and women: the Dubbo Osteoporosis Epidemiology Study (DOES). Osteoporos Int 4:277282.
73
243. Roerholt C, Eiken P, Abrahamsen B 2/2009 Initiation of anti-osteoporotic therapy in patients with recent
fractures: a nationwide analysis of prescription rates and persistence. Osteoporos Int 20:299-307.
244. Bischoff-Ferrari HA, Willett WC, Wong JB, Giovannucci E, Dietrich T, wson-Hughes B 5/11/2005 Fracture
prevention with vitamin D supplementation: a meta-analysis of randomized controlled trials. JAMA
293:2257-2264.
245. Boonen S, Lips P, Bouillon R, Bischoff-Ferrari HA, Vanderschueren D, Haentjens P 4/2007 Need for additional
calcium to reduce the risk of hip fracture with vitamin d supplementation: evidence from a comparative
metaanalysis of randomized controlled trials. J Clin Endocrinol Metab 92:1415-1423.
246. Orwoll E, Ettinger M, Weiss S, Miller P, Kendler D, Graham J, Adami S, Weber K, Lorenc R, Pietschmann P,
Vandormael K, Lombardi A 8/31/2000 Alendronate for the treatment of osteoporosis in men. N Engl J
Med 343:604-610.
247. Ringe JD, Faber H, Dorst A 11/2001 Alendronate treatment of established primary osteoporosis in men: results of
a 2-year prospective study. J Clin Endocrinol Metab 86:5252-5255.
248. Gonnelli S, Cepollaro C, Montagnani A, Bruni D, Caffarelli C, Breschi M, Gennari L, Gennari C, Nuti R 8/2003
Alendronate treatment in men with primary osteoporosis: a three-year longitudinal study. Calcif Tissue
Int 73:133-139.
249. Boonen S, Orwoll ES, Wenderoth D, Stoner KJ, Eusebio R, Delmas PD 4/2009 Once-weekly risedronate in men
with osteoporosis: results of a 2-year, placebo-controlled, double-blind, multicenter study. J Bone Miner
Res 24:719-725.
250. Orwoll ES, Binkley NC, Lewiecki EM, Gruntmanis U, Fries MA, Dasic G 4/2010 Efficacy and safety of monthly
ibandronate in men with low bone density. Bone 46:970-976.
251. Fahrleitner-Pammer A, Piswanger-Soelkner JC, Pieber TR, Obermayer-Pietsch BM, Pilz S, Dimai HP, Prenner G,
Tscheliessnigg KH, Hauge E, Portugaller RH, Dobnig H 7/2009 Ibandronate prevents bone loss and
reduces vertebral fracture risk in male cardiac transplant patients: a randomized double-blind, placebocontrolled trial. J Bone Miner Res 24:1335-1344.
252. Lyles KW, Colon-Emeric CS, Magaziner JS, Adachi JD, Pieper CF, Mautalen C, Hyldstrup L, Recknor C,
Nordsletten L, Moore KA, Lavecchia C, Zhang J, Mesenbrink P, Hodgson PK, Abrams K, Orloff JJ,
Horowitz Z, Eriksen EF, Boonen S 11/1/2007 Zoledronic acid and clinical fractures and mortality after
hip fracture. N Engl J Med 357:1799-1809.
253. Orwoll ES, Scheele WH, Paul S, Adami S, Syversen U, ez-Perez A, Kaufman JM, Clancy AD, Gaich GA 1/2003
The effect of teriparatide [human parathyroid hormone (1-34)] therapy on bone density in men with
osteoporosis. J Bone Miner Res 18:9-17.
254. Kurland ES, Cosman F, McMahon DJ, Rosen CJ, Lindsay R, Bilezikian JP 9/2000 Parathyroid hormone as a
therapy for idiopathic osteoporosis in men: effects on bone mineral density and bone markers. J Clin
Endocrinol Metab 85:3069-3076.
255. Kurland ES, Heller SL, Diamond B, McMahon DJ, Cosman F, Bilezikian JP 12/2004 The importance of
bisphosphonate therapy in maintaining bone mass in men after therapy with teriparatide [human
parathyroid hormone(1-34)]. Osteoporos Int 15:992-997.
256. Finkelstein JS, Hayes A, Hunzelman JL, Wyland JJ, Lee H, Neer RM 9/25/2003 The effects of parathyroid
hormone, alendronate, or both in men with osteoporosis. N Engl J Med 349:1216-1226.
257. Behre HM, Kliesch S, Leifke E, Link TM, Nieschlag E 8/1997 Long-term effect of testosterone therapy on bone
mineral density in hypogonadal men. J Clin Endocrinol Metab 82:2386-2390.
74
258. Snyder PJ, Peachey H, Hannoush P, Berlin JA, Loh L, Holmes JH, Dlewati A, Staley J, Santanna J, Kapoor SC,
Attie MF, Haddad JG, Jr., Strom BL 6/1999 Effect of testosterone treatment on bone mineral density in
men over 65 years of age. J Clin Endocrinol Metab 84:1966-1972.
259. Kenny AM, Prestwood KM, Gruman CA, Marcello KM, Raisz LG 5/2001 Effects of transdermal testosterone on
bone and muscle in older men with low bioavailable testosterone levels. J Gerontol A Biol Sci Med Sci
56:M266-M272.
260. Amory JK, Watts NB, Easley KA, Sutton PR, Anawalt BD, Matsumoto AM, Bremner WJ, Tenover JL 2/2004
Exogenous testosterone or testosterone with finasteride increases bone mineral density in older men with
low serum testosterone. J Clin Endocrinol Metab 89:503-510.
261. Doran PM, Riggs BL, Atkinson EJ, Khosla S 11/2001 Effects of raloxifene, a selective estrogen receptor
modulator, on bone turnover markers and serum sex steroid and lipid levels in elderly men. J Bone Miner
Res 16:2118-2125.
262. Trovas GP, Lyritis GP, Galanos A, Raptou P, Constantelou E 3/2002 A randomized trial of nasal spray salmon
calcitonin in men with idiopathic osteoporosis: effects on bone mineral density and bone markers. J Bone
Miner Res 17:521-527.
263. Saag KG, Emkey R, Schnitzer TJ, Brown JP, Hawkins F, Goemaere S, Thamsborg G, Liberman UA, Delmas PD,
Malice MP, Czachur M, Daifotis AG 7/30/1998 Alendronate for the prevention and treatment of
glucocorticoid-induced osteoporosis. Glucocorticoid-Induced Osteoporosis Intervention Study Group. N
Engl J Med 339:292-299.
264. Wallach S, Cohen S, Reid DM, Hughes RA, Hosking DJ, Laan RF, Doherty SM, Maricic M, Rosen C, Brown J,
Barton I, Chines AA 10/2000 Effects of risedronate treatment on bone density and vertebral fracture in
patients on corticosteroid therapy. Calcif Tissue Int 67:277-285.
265. Ringe JD, Dorst A, Faber H, Ibach K, Preuss J 6/2003 Three-monthly ibandronate bolus injection offers
favourable tolerability and sustained efficacy advantage over two years in established corticosteroidinduced osteoporosis. Rheumatology (Oxford) 42:743-749.
266. Reid DM, Devogelaer JP, Saag K, Roux C, Lau CS, Reginster JY, Papanastasiou P, Ferreira A, Hartl F, Fashola T,
Mesenbrink P, Sambrook PN 4/11/2009 Zoledronic acid and risedronate in the prevention and
treatment of glucocorticoid-induced osteoporosis (HORIZON): a multicentre, double-blind, doubledummy, randomised controlled trial. Lancet 373:1253-1263.
267. Saag KG, Shane E, Boonen S, Marin F, Donley DW, Taylor KA, Dalsky GP, Marcus R 11/15/2007 Teriparatide
or alendronate in glucocorticoid-induced osteoporosis. N Engl J Med 357:2028-2039.
268. Langdahl BL, Marin F, Shane E, Dobnig H, Zanchetta JR, Maricic M, Krohn K, See K, Warner MR 12/2009
Teriparatide versus alendronate for treating glucocorticoid-induced osteoporosis: an analysis by gender
and menopausal status. Osteoporos Int 20:2095-2104.
269. Greenspan SL, Nelson JB, Trump DL, Wagner JM, Miller ME, Perera S, Resnick NM 9/20/2008 Skeletal health
after continuation, withdrawal, or delay of alendronate in men with prostate cancer undergoing androgendeprivation therapy. J Clin Oncol 26:4426-4434.
270. Satoh T, Kimura M, Matsumoto K, Tabata K, Okusa H, Bessho H, Iwamura M, Ishiyama H, Hayakawa K, Baba S
8/1/2009 Single infusion of zoledronic acid to prevent androgen deprivation therapy-induced bone loss
in men with hormone-naive prostate carcinoma. Cancer 115:3468-3474.
271. Smith MR, McGovern FJ, Zietman AL, Fallon MA, Hayden DL, Schoenfeld DA, Kantoff PW, Finkelstein JS
9/27/2001 Pamidronate to prevent bone loss during androgen-deprivation therapy for prostate cancer. N
Engl J Med 345:948-955.
75
272. Smith MR, Eastham J, Gleason DM, Shasha D, Tchekmedyian S, Zinner N 6/2003 Randomized controlled trial of
zoledronic acid to prevent bone loss in men receiving androgen deprivation therapy for nonmetastatic
prostate cancer. J Urol 169:2008-2012.
273. Michaelson MD, Kaufman DS, Lee H, McGovern FJ, Kantoff PW, Fallon MA, Finkelstein JS, Smith MR
3/20/2007 Randomized controlled trial of annual zoledronic acid to prevent gonadotropin-releasing
hormone agonist-induced bone loss in men with prostate cancer. J Clin Oncol 25:1038-1042.
274. Smith MR, Egerdie B, Hernandez TN, Feldman R, Tammela TL, Saad F, Heracek J, Szwedowski M, Ke C, Kupic
A, Leder BZ, Goessl C 8/20/2009 Denosumab in men receiving androgen-deprivation therapy for
prostate cancer. N Engl J Med 361:745-755.
275. Smith MR, Fallon MA, Lee H, Finkelstein JS 8/2004 Raloxifene to prevent gonadotropin-releasing hormone
agonist-induced bone loss in men with prostate cancer: a randomized controlled trial. J Clin Endocrinol
Metab 89:3841-3846.
276. Smith MR, Morton RA, Barnette KG, Sieber PR, Malkowicz SB, Rodriguez D, Hancock ML, Steiner MS
8/17/2010 Toremifene to Reduce Fracture Risk in Men Receiving Androgen Deprivation Therapy for
Prostate Cancer. J Urol.
277. Hwang JS, Liou MJ, Ho C, Lin JD, Huang YY, Wang CJ, Tsai KS, Chen JF 12/15/2009 The effects of weekly
alendronate therapy in Taiwanese males with osteoporosis. J Bone Miner Metab.
278. de Nijs RN, Jacobs JW, Lems WF, Laan RF, Algra A, Huisman AM, Buskens E, De Laet CE, Oostveen AC,
Geusens PP, Bruyn GA, Dijkmans BA, Bijlsma JW 8/17/2006 Alendronate or alfacalcidol in
glucocorticoid-induced osteoporosis. N Engl J Med 355:675-684.
279. Orwoll ES, Miller PD, Adachi JD, Brown J, Adler RA, Kendler D, Bucci-Rechtweg C, Readie A, Mesenbrink P,
Weinstein RS 10/2010 Efficacy and safety of a once-yearly i.v. Infusion of zoledronic acid 5 mg versus a
once-weekly 70-mg oral alendronate in the treatment of male osteoporosis: A randomized, multicenter,
double-blind, active-controlled study. J Bone Miner Res 25:2239-2250.
280. Finkelstein JS, Hayes A, Hunzelman JL, Wyland JJ, Lee H, Neer RM 9/25/2003 The effects of parathyroid
hormone, alendronate, or both in men with osteoporosis. N Engl J Med 349:1216-1226.
281. Smith MR, Saad F, Egerdie B, Szwedowski M, Tammela TL, Ke C, Leder BZ, Goessl C 12/2009 Effects of
denosumab on bone mineral density in men receiving androgen deprivation therapy for prostate cancer. J
Urol 182:2670-2675.
282. Emmelot-Vonk MH, Verhaar HJ, Nakhai Pour HR, Aleman A, Lock TM, Bosch JL, Grobbee DE, van der
Schouw YT 1/2/2008 Effect of testosterone supplementation on functional mobility, cognition, and
other parameters in older men: a randomized controlled trial. JAMA 299:39-52.
283. National Osteoporosis Guideline Group 2010 Guideline for the diagnosis and management of osteoporosis.
284. Brown JP, Josse RG 11/12/2002 2002 clinical practice guidelines for the diagnosis and management of
osteoporosis in Canada. CMAJ 167:S1-34.
285. Qaseem A, Snow V, Shekelle P, Hopkins R, Jr., Forciea MA, Owens DK 5/6/2008 Screening for osteoporosis in
men: a clinical practice guideline from the American College of Physicians. Ann Intern Med 148:680-684.
286. Nelson HD, Helfand M, Woolf SH, Allan JD 9/17/2002 Screening for postmenopausal osteoporosis: a review of
the evidence for the U.S. Preventive Services Task Force. Ann Intern Med 137:529-541.
76
21 Vitamin D status and PTH in young men: a cross-sectional study on associations with bone
mineral density, body composition and glucose metabolism
Frost M1, Abrahamsen B2, Nielsen TL1, Hagen C3, Andersen M1, and, Brixen K1
1
Department of Endocrinology, Odense University Hospital, Odense, Denmark
2
Department of Internal Medicine, Copenhagen University Hospital Gentofte, Copenhagen,
Denmark
3
.
Department of Endocrinology, Bispebjerg Hospital, Copenhagen, Denmark
77
Abstract:
Objective: Though vitamin D and bone metabolism are closely related, few studies have addressed
the effects of vitamin D status on bone in men at time of peak bone mass. The objectives of this
study were to evaluate the prevalence of vitamin D inadequacy in a cross-sectional study on young
men and the effects of vitamin D and parathyroid hormone (PTH) on bone mass, bone markers and
metabolic function.
Design and Participants: The study population consisted of 783 men aged 20-29 yrs.
Measurements: Bone mineral density (BMD) of the total hip, femoral neck and lumbar spine were
measured. DXA was used to evaluate total body fat mass (BFAT). Visceral and abdominal
subcutaneous fat mass (ViFM and ScFM) were assessed by use of magnetic resonance imaging. A
radioimmunoassay was used to measure the level of 25-hydroxy vitamin D (25OHD).
Results: The prevalence of vitamin deficiency (serum 25OHD<50nM) was 6.3% during summer
and 43.6% during winter. Serum 25OHD was associated with BMD at all sites and inversely
associated with bone-specific alkaline phosphatase and directly with carboxyterminal telopeptide of
type-1-collagen. 25OHD and PTH were inversely associated with BFAT, whereas 25OHD also was
inversely associated with body mass index, waist-hip ratio, ViFM and ScFM after adjustment for
confounders. The associations were found only to be present in participants with non-sufficient
levels of 25OHD. 25OHD and PTH were inversely related to insulin resistance in vitamin
insufficient participants only. No associations between PTH or 25OHD and blood pressures were
noted.
Conclusion: The study showed a high prevalence of 25OHD deficiency in young, Northern
European men significantly associated with BMD. PTH and 25OHD were found to be inversely
related to markers of insulin resistance.
78
Introduction
Vitamin D and parathyroid hormone (PTH) are central to the regulation of calcium and bone
metabolism. Studies have shown association between serum levels of vitamin D and bone mass (1)
and an inverse relation with the risk of osteoporotic fractures (2), moreover some (3) but not all
studies (4) show fracture reduction on treatment with vitamin D and calcium. PTH is secreted in
response to reduced calcium levels causing an increase in bone resorption and subsequently
normalisation of calcium levels. In case of vitamin D deficiency, secondary hyperparathyroidism
resulting in increased bone turnover and increased bone loss and “remodelling space” may
subsequently increase fracture risk (2).
Peak bone mass (PBM) is a determinant of bone health in later life (5). The importance of PBM is
emphasized by a doubling of fracture risk for every decline of one standard deviation of bone
mineral density (BMD) (6). Any factor favouring gain of bone mass in early adulthood is, therefore,
of great significance in order to improve bone health in late life.
Most of the variation in PBM is explained by genetics (7), but several factors known to influence
PBM are modifiable including physical activity (8) and levels of vitamin D (10). Studies have
linked vitamin D to non-skeletal disorders including type I diabetes and cancers (11) as well as
obesity (12), insulin secretion (13), hypertension and metabolic syndrome (14). PTH itself has been
related to both hypertension (15) and the metabolic syndrome (16) independently of vitamin D.
The aims of the present study were first to investigate the prevalence of vitamin D deficiency and
insufficiency in young men at the time of peak bone mass. The definition of vitamin D status used
was proposed in a recent report (17); vitamin D deficiency defined as levels of 25OHD below 50
nmol/l and insufficiency as levels lower than 75 nmol/l. Second, to assess the impact of modifiable
predictors including lifestyle factors and body composition on vitamin D status. Third, to evaluate
the relationship between vitamin D and PBM, bone markers and markers of metabolic syndrome.
79
Subjects and methods
Subjects
This is a population-based observational study on endocrine status, body composition, and bone
metabolism in young men conducted in the Danish city Odense (55.2°N).
The participants were recruited from a random selection of 3,000 men aged 20-29
years who all received a questionnaire concerning medical history, medication and lifestyle factors.
Seventy-three per cent responded and were invited to participate in the study; in all, 783 Caucasian
men consented and underwent medical examinations as well as an interview about issues affecting
bone health i.e. physical activity, smoking, alcohol consumption, and vitamin D supplementation.
All cases of bone disease (n=1), anabolic steroids (n=19), systemic corticosteroids
(n=7), thyroid disorders (n=10), low LH (<0.1; n=6), bilateral small testicles (volume <9ml; n=16),
or excessive alcohol intake (>6U/day, n=18) were excluded. Participants with a chronic disease
affecting bone and vitamin D levels were non-eligible (celiac disease n=2, alcoholism n=1,
inflammatory bowel disease n=2, lactose intolerance n=2, diabetes mellitus n=2, Marfan’s
syndrome n=1, epilepsy n=5, congenital cretinism n=1, chronic hepatitis n=1, glomerulonephritis
n=1, ankylosing spondylitis n=2). In all, 700 men were included in the study.
The study was approved by the Local Ethics Committee (file 20010198) and
registered in ClinicalTrials.gov as NTC00150163. Participants consented in writing, and the study
was carried out in accordance with the Helsinki II declaration.
Bone densitometry
Dual energy X-ray absorptiometry (DXA) (H4500, Hologic Inc., Waltham, MA) was used to
measure BMD (g/cm2) in the lumbar spine (L2-L4), hip and femoral neck. The coefficients of
variation (CV) were 1.5% for both the lumbar spine and total hip measurements.
80
Body composition
Waist and hip circumference (cm) were measured at the navel and at the greater trochanter, and
waist-hip ratio (WHR) was calculated as waist divided by hip circumference. Body mass index
(BMI, Kg/m2) was calculated as weight (SECA, Hamburg, Germany) divided by height2
(Harpenden, Holtain Ltd, Crymmych, UK). Lean body mass (LBM) and total body fat (BFAT) were
measured by DXA. CV’s of LBM and BFAT were 3% and 3.1%, respectively.
Magnetic resonance imaging (MRI) (Siemens, Erlangen, Germany) was used to evaluate visceral
and abdominal subcutaneous fat mass (ViFM and ScFM, respectively) in the first 364 consecutive
participants. . Abdominal subcutaneous and visceral fat mass was assessed by evaluating 3 slices
recorded at the intervertebral space of L4/L5 (CV: subcutaneous: 1.7%. visceral: 7.2%)
Lifestyle and blood pressure
Smoking status was filed dichotomously as never/previous and current smoker. Intake of alcohol
was noted as weekly consumption in units (one unit equalling 12 grams of alcohol). Use of
multivitamin supplement was registered as yes or no. Consumption of vitamin D was otherwise not
registered, however, in Denmark, food is with the exception of margarine not fortified with vitamin
D. Physical activity was recorded as to hours spent per week on sports. Blood pressure was
measured twice in a sitting position after 5 minutes of relaxation.
Biochemistry
All samples were collected between 8 and 10 a.m. after an overnight fast of at least 8 hours. Serum
25OHD was measured using a radioimmunoassay (DiaSorin, Stillwater, MN) (intra-assay CV: 6%).
Vitamin D deficiency was defined as levels of 25OHD below 50 nmol/l and insufficiency as levels
81
lower than 75 nmol/l. A immunometric assay (Immunolite 2000, DPC, Los Angeles, CA) was used
to measure serum PTH (CV: 5%). Serum TSH was measured with a non-immunoflourometric assay
(CV: 2%), and a non-competitive immunoflourometric assay was used to measure serum LH (CV:
5%) (both: Delfia, WALLAC, Turku, Finland). Bone specific alkaline phosphatase (BAP) was
measured by use of monoclonal capture (Metra Biosystems Inc., Mountain View, CA) (CV: 4%)
while a RIA assay (Orion Diagnostica, Espoo, Finland) was used to evaluate the bone marker type1 collagen C-terminal telopeptide (1CTP) (CV: 4%).
Insulin was measured using a non-competitive immunoflourometric assay (Delfia, WALLAC,
Turku, Finland) (CV: 3%) and plasma glucose by use of a hexokinase-based method (Roche,
Mannheim, Germany) (CV: 5.7%). Homeostasis model assessment of insulin resistance (HOMA
(IR)) was calculated using the formula: s-insulin (pmol/l) x plasma glucose divided by 162 (18).
Statistics
Data are shown as mean ± standard deviation or median [25-75 percentiles] when appropriate.
Parameters with non-normal distribution according to distribution plots and the Shapiro-Wilk’s test
were log-transformed (The distributions of insulin, HOMA(IR), PTH, BAP, 1CTP, ViFM, ScFM
and BFAT but not alcohol were rectified), and statistical calculations were performed on logtransformed values. Locally weighted linear regression (Lowess) was used to depict the relationship
between serum 25OHD and measures of bone mass, levels of PTH, bone markers and BFAT
graphically. Pearson’s correlation coefficients for correlations between 25OHD as well as PTH
(log-transformed) and continuous variables were calculated. ANOVA and chi-square tests were
used to compare continuous and categorical outcomes, respectively, between groups defined by
vitamin D status. Kruskall-Wallis test was used for evaluation of use of alcohol.
82
Determinants of vitamin D sufficiency were evaluated by use of logistic regression analysis
including vitamin D status as a dichotomous dependent variable and age, season, multivitamin
supplements, smoking, alcohol, sports and BMI as independent variables. Multiple regression
analyses were performed with 25OHD as the dependent variable and first BMI and secondly BFAT
and LBM as independent variables. Moreover, multiple regression analyses were applied in the
investigation of association between 25OHD as well as PTH and markers of metabolic dysfunction
including blood pressure, body composition (waist-hip ratio, LBM, BFAT, ViFM and ScFM), lipids
and insulin resistance as dependent variables. The residuals were found to be normally distributed
according to normality plot.
All statistics were performed with STATA v10 (STATA, TX, USA). P-values less than 5% were
considered significant.
Results
General characteristics
Median age of the participants was 25.7 [range: 19.4-30.9] years. The study population had a BMI
of 24.7 (3.4) kg/m2). A total of 32.9 % (n=230) were smokers, 13 % (n=89) reported taking
multivitamins (containing 200 IU of vitamin D) and 46 % (n=322) were actively participating in
sports.
Vitamin D status
The overall serum 25OHD was 64.9 (27.7) nmol/L, and the prevalence of vitamin D insufficiency
and deficiency were 31.4% (n=220) and 32.7% (n=229), respectively. As presented in table 1, the
prevalence of vitamin-D deficiency was 43.6 % during winter and 6.3 % in the summer (table 1).
83
Corresponding levels of serum 25OHD in winter and summer were 59.3 (27.1) and 75.0 (19.6)
nmol/l, respectively.
A larger part of vitamin D sufficient participants were taking multivitamin supplements compared
to insufficient or deficient participants (18 % vs. 12 % and 8 %, respectively. p=0.002) and they
participated more often in sports (57 % vs. 43 % and 37 %, respectively. p<0.001), whereas fewer
were smokers (28 % vs. 32 % and 40 %, p=0.015) (table 2). No significant difference in alcohol
consumption was found.
Calcium homeostasis
Vitamin D insufficient participants had significantly higher serum levels of PTH, whereas the level
of ionized calcium was unrelated to vitamin D status (table 2). Serum BAP was increased in vitamin
D deficient participants (table 2 and fig 1) while the serum level of 1CTP was higher in vitamin D
sufficient and deficient participants. BAP remained inversely associated with 25OHD in nonsufficient participants, whereas 1CTP was non-significantly associated with 25OHD after
stratification according to vitamin D status.
25OHD was independently and inversely associated with PTH (β = -0.17, p<0.001) as well as BAP
(β = -0.12, p=0.015) in the vitamin D insufficient participants only.
While PTH was inversely associated with calcium levels irrespective of vitamin D status, 25OHD
was positively associated with the level of calcium in sufficient and inversely associated with
calcium in insufficient cases (table 3).
Bone mass
Participants with a sufficient level of vitamin D had significantly higher levels of BMD at all sites
investigated. For the femoral neck, total hip and lumbar spine, BMD was 4.3 % (0.96 (95%CI:
84
0.99-0.17) vs. 0.92 (0.96-1.13) g/cm2, p=0.004), 3.8% (1.09 (0.98-1.18) vs. 1.05 (0.96-1.14) g/cm2,
p=0.032) and 1.9% (1.08 (0.99-1.17) vs. 1.06 (0.96-1.13) g/cm2, p=0.024) higher compared to
vitamin D deficient participants, respectively. The relations between vitamin D levels and BMD are
presented as Lowess plots in figures 1-2. BMD appears to be stable at levels of 25OHD of at least
50nM and upwards.
BMD and serum 25-OHD were unrelated in participants with 25-OHD-levels above 75 nM,
whereas 25OHD was significantly associated with BMDlumbar (β =0.17), BMD neck (β =0.19), and
BMD total hip (β =0.24, all p<0.001) in those with a serum 25OHD below 75nM (Table 3).
Body composition
BMI, waist/hip ratio, BFAT and visceral as well as subcutaneous FM were significantly different
among groups defined by vitamin D status (table 2). All measures of FM were inversely correlated
with serum 25OHD in unadjusted (r= -0.10; -0.22) and adjusted analyses (β= -0.11; -0.17). In
separate analyses, 25OHD was found to be inversely associated with measures of FM in vitamin D
non-sufficient participants only (β= -0.12; -0.18) (table 3).
An inverse relationship between 25OHD and BMI was present in participants with a high BMI (fig
3); those with a BMI above 25 Kg/m2 (n=244) had a lower level of 25OHD (61.4 (27.8) vs. 66.7
(27.5) nmol/l, p=0.015). In participants with a BMI exceeding 25 Kg/m2, an increase in BMI of 1
Kg/m2 corresponded to a decrease in 25OHD of 1.7 nM (95%CI: -2.8; -0.6, p=0.002) whereas BMI
and 25OHD was unrelated in participants with a BMI below 25 Kg/m2 (coef: 0.7 (95%CI: -0.6; 2.0)).
PTH was found to be inversely associated with BFAT (β = -0.11) in vitamin D insufficient persons
only. Additionally, 25OHD but not PTH was associated with LBM in adjusted analyses and in
vitamin D insufficient participants specifically (table 3).
85
Insulin resistance, lipids and blood pressure
The plasma level of insulin was significantly higher in participants with vitamin D deficiency,
whereas the level of plasma glucose was similar in all groups. Accordingly, HOMA(IR) was higher
in vitamin D deficient participants. Serum levels of triglycerides and HDL but not LDL were
significantly different between groups defined by vitamin D status (all p<0.01) (table 2).
In unadjusted analyses, PTH and 25OHD inversely associated with insulin, whereas the latter in
addition was negatively associated with HOMA(IR), LDL and triglycerides and directly associated
with HDL (table 3). After stratification according to vitamin D status, serum 25OHD and PTH were
inversely associated with insulin and HOMA(IR) in non-sufficient participants only (table 3). PTH
was inversely associated with the level of triglycerides in non-sufficient participants and directly
associated with HDL in sufficient participants.
Neither unadjusted nor adjusted analyses, suggested any association between PTH or 25OHD and
blood pressure (table 3). Stratification according to vitamin D sufficiency had no impact on the
coefficients.
Predictors of vitamin D sufficiency
Using logistic regression, the likelihood of being vitamin D sufficient was higher in those taking
multivitamin supplementation (OR: 2.40 (95%CI: 1.45-3.84). p=0.001), study participation in
seasons “Summer” and “Autumn”, (OR: 3.67 (95%CI: 2.23-6.03) and OR: 1.96 (95%CI: 1.312.92)) as well as partaking in sports (OR: 1.10 (95%CI: 1.05-1.14)) (all p<0.001).
As a continuous variable, serum 25OHD was in multiple regression analysis associated with age
(β=0.09; p=0.02), sports (r=0.17: p<0.001), taking multivitamins (β=0.11; p=0.003) as well as
86
seasons “Summer” and “Autumn” (β=0.23 and β=0.19; both p<0.001). Smoking and BMI were
inversely associated with lower levels of serum 25OHD (β=-0.09; p=0.020 and β=-0.08; p=0.021).
Median 25OHD in persons taking multivitamins was 72.8 (23.1) nmol/l and in non-users 63.7
(28.1) nmol/l (p=0.004). In the multiple regression analyses, taking multivitamins was associated
with an increase in 25OHD of 9.1 nmol/L.
With BFAT and LBM rather than BMI in the model, BFAT was found to be inversely (β=-0.21;
p<0.001) and LBM directly associated with 25OHD (β=0.13; p=0.004) (R2adj =0.13). In separate
multiple regression analyses, both visceral FM (β: -0.13; p=0.017) (R2adj =0.13) and subcutaneous
FM (β: -0.14; p=0.013) (R2adj =0.13) were inversely associated with 25OHD.
Discussion
This study showed seasonal variation in 25OHD and a high prevalence of vitamin D deficiency in
young northern-European men reaching 43.6 % during winter. Vitamin D insufficiency was,
however, also prevalent during summer (41.1 %) and fall (38.3 %). Although the average 25OHD
level was higher in summer, the increase was inadequate translating into a lower number of
deficient and a higher number of insufficient individuals. These results emphasize the need of
improvement of vitamin D status among otherwise healthy young men.
Another major finding was a reduced level of BMD at all sites investigated at serum 25OHD levels
below 50 nM. In absolute terms, the difference in BMD in the lumbar spine, femoral neck and total
hip between vitamin D deficient and sufficient participants was 2-3 % corresponding to 0.15 SD for
the lumbar spine and 0.3S D for the femoral neck and total hip. If the difference would remain into
late adulthood, this could equal an increase in vertebral fracture risk of 35 % and hip fracture of 80
% (6).
87
Few studies have addressed the effects of vitamin D status on bone in men at time of PBM. During
wintertime, 70 % of French boys aged 13-16 yrs were found to be vitamin D deficient (19) and in
Finland, a study on 220 primarily army-recruits aged 18.3-20.6 yrs showed a prevalence of vitamin
D deficiency of approximately 40 % during winter and reduced levels of BMD in the lumbar spine
and total hip were reported in vitamin D deficient participants. Additionally, in 78 Swedish men
mean age 22.6 yrs, cholecalciferol was associated with total body and lumbar spine BMD (20). In
elderly men, 25OHD has been associated with BMD at the hip and spine (21) and increased risk of
hip fractures (22). In women, studies have found both absence of an association between 25OHD
and BMD in Icelandic women aged 16-20 (23) and an association between the same parameters in
14-16 yrs old Finnish adolescents (24). The NHANES III data showed an association between
serum 25-OHD and BMD irrespective of gender, race and age (25), although recent data showed
effect of vitamin D deficiency on BMD in white men primarily (26).
Our study population consisted of men aged 20-29 yrs. Although PBM is largely attained by the end
of the second decade, bone mineral accrual may continue into the third decade (27) suggesting that
our study population would have reached PBM. In a 8-year follow-up study on BMD in men aged
17-26 yrs, Nordström et al. (28) found loss of BMD in the hip from the age of 19 yrs but even
levels of total body and lumbar spine BMD during the last part of the study period. Changes in body
composition may, however, influence the measurement of PBM as increases in lumbar spine BMD
measured by DXA would continue even though increases in bone mass measured by quantitative
computed tomography had ceased (29).
In the present study, BMD levels appeared to increase until 25OHD reached a level of 70-80 nmol/l.
Considerable increases in both PTH and BAP were found at levels of 25OHD below 50 nmol/l
suggesting secondary hyperparathyroidism may be prevalent in the study population. Although
participants were young, secondary hyperparathyroidism could have deleterious effects on future
88
fracture risk at this age (30). Three of the participants had 25OHD below 15nM. Should these cases
indeed have osteomalacia, the overall impact on the results is probably small.
Consistent with other studies, we found an inverse association between 25OHD and BMI, WHR as
well as BFAT and ViFM. On stratification according to vitamin D status, the associations persisted
only in non-sufficient participants even after adjustment for life style factors and seasonal variation.
Previous studies (31;32) have shown inverse associations between 25OHD and waist-hip ratio or fat
mass as well as a more close association between 25OHD and visceral FM than to subcutaneous fat
mass (33). Although the association between visceral FM and 25OHD appeared stronger than that
of subcutaneous FM and 25OHD, the correlation coefficients were comparable, and we cannot
conclude if the distribution of fat mass is related to vitamin D.
In our study, 25OHD was negatively associated with HOMA(IR) and triglycerides independently of
BMI and several life style factors, where as no adjusted associations were found with HDL and
LDL. Interestingly, the significant findings were made in non-sufficient participants only. The
unfavourable relationships between 25OHD and markers of metabolic dysfunction, i.e. HOMA(IR),
may be present at 25OHD below certain levels. Some earlier studies have found inverse association
between 25OHD and insulin sensitivity or metabolic syndrome (32), whereas others have not (16).
Accordingly, Pitas et al. found a positive effect of vitamin D on glucose metabolism in a placebocontrolled 3 yr trial (34).
In some (35) but not all (15) epidemiological studies, vitamin D and blood pressure are inversely
associated. The effect of vitamin D on blood pressure is not evident, and randomised trials are
needed for clarification. Snijder et al. (15) found that high levels PTH was associated with
increased systolic and diastolic blood pressure in a large population-based Dutch study. Our data
did not suggest any association between vitamin D or PTH and blood pressure; perhaps due to
differences in age of the participants.
89
The likelihood of vitamin D sufficiency was increased if multivitamins were used. Even though
25OHD was higher in users, the level remained sub-optimal. Since the brand of the multivitamins
was not registered, we cannot account for the use of either D2 or D3 possibly influencing the
evaluation of effects of intake of multivitamins Partaking in sports and season “summer” and
“autumn” were also associated with increased 25OHD. These variables may include exposure to
sunlight and information on lifestyle factors not accounted for in this study, i.e. food intake. Data
also showed an inverse association between smoking as well as obesity on 25OHD levels. These
predictors of sufficient vitamin D levels could be targeted in interventions for improvement vitamin
D status.
A strength of our study was random recruitment of the study participants from the back-ground
population. Additionally, the study population was homogenous and consisted of Caucasians living
in a Northern-European country only. On the other hand, our results may not be applicable to other
populations. Another shortcoming was the cross-sectional design that deters us from evaluating the
direction of cause and effect. In addition, several tests are performed increasing the risk of type I
errors. Nevertheless, we believe data could be used for novel hypotheses.
In summery, we found a high prevalence of inadequate levels of vitamin D in a population-based
study on young men at time of PBM. We also found higher levels of PTH and BAP as well as lower
PBM in participants with inadequate vitamin D status. Whether vitamin D supplements would
increase levels of BMD at this time of life remains to be tested, however, should that be the case,
future fracture risk could potentially be considerably reduced by increasing levels of 25OHD in this
age group.
Additionally, we found inverse associations between vitamin D and markers of metabolic functions
in participants with inappropriate vitamin D status. Whether increased vitamin D intake indeed
reduces the incidence of obesity or improve glucose metabolism should be tested in the future.
90
Acknowledgements:
Financial support was received from Velux Foundation, WADA, Novo Nordisk, Ministry of
Culture and Clinical Institute at the University of Southern Denmark.
91
Reference List
(1) Valimaki VV, Alfthan H, Lehmuskallio E, Loyttyniemi E, Sahi T, Stenman UH, et al. Vitamin D status as a
determinant of peak bone mass in young Finnish men. J Clin Endocrinol Metab 2004 Jan;89(1):76-80.
(2) Lips P. Vitamin D deficiency and secondary hyperparathyroidism in the elderly: consequences for bone loss and
fractures and therapeutic implications. Endocr Rev 2001 Aug;22(4):477-501.
(3) Patient level pooled analysis of 68 500 patients from seven major vitamin D fracture trials in US and Europe. BMJ
2010;340:b5463.
(4) Grant AM, Avenell A, Campbell MK, McDonald AM, MacLennan GS, McPherson GC, et al. Oral vitamin D3 and
calcium for secondary prevention of low-trauma fractures in elderly people (Randomised Evaluation of Calcium
Or vitamin D, RECORD): a randomised placebo-controlled trial1. Lancet 2005 May 7;365(9471):1621-8.
(5) Eisman JA, Kelly PJ, Morrison NA, Pocock NA, Yeoman R, Birmingham J, et al. Peak bone mass and
osteoporosis prevention. Osteoporos Int 1993;3 Suppl 1:56-60.
(6) Marshall D, Johnell O, Wedel H. Meta-analysis of how well measures of bone mineral density predict occurrence
of osteoporotic fractures. BMJ 1996 May 18;312(7041):1254-9.
(7) Hopper JL, Green RM, Nowson CA, Young D, Sherwin AJ, Kaymakci B, et al. Genetic, common environment,
and individual specific components of variance for bone mineral density in 10- to 26-year-old females: a twin
study. Am J Epidemiol 1998 Jan 1;147(1):17-29.
(8) Hind K, Burrows M. Weight-bearing exercise and bone mineral accrual in children and adolescents: a review of
controlled trials. Bone 2007 Jan;40(1):14-27.
(9) Winzenberg T, Shaw K, Fryer J, Jones G. Effects of calcium supplementation on bone density in healthy children:
meta-analysis of randomised controlled trials. BMJ 2006 Oct 14;333(7572):775.
(10) Kremer R, Campbell PP, Reinhardt T, Gilsanz V. Vitamin D status and its relationship to body fat, final height,
and peak bone mass in young women. J Clin Endocrinol Metab 2009 Jan;94(1):67-73.
(11) Holick MF. Vitamin D deficiency. N Engl J Med 2007 Jul 19;357(3):266-81.
(12) Snijder MB, van Dam RM, Visser M, Deeg DJ, Dekker JM, Bouter LM, et al. Adiposity in relation to vitamin D
status and parathyroid hormone levels: a population-based study in older men and women. J Clin Endocrinol
Metab 2005 Jul;90(7):4119-23.
(13) Norman AW, Frankel JB, Heldt AM, Grodsky GM. Vitamin D deficiency inhibits pancreatic secretion of insulin.
Science 1980 Aug 15;209(4458):823-5.
(14) Martins D, Wolf M, Pan D, Zadshir A, Tareen N, Thadhani R, et al. Prevalence of cardiovascular risk factors and
the serum levels of 25-hydroxyvitamin D in the United States: data from the Third National Health and Nutrition
Examination Survey. Arch Intern Med 2007 Jun 11;167(11):1159-65.
(15) Snijder MB, Lips P, Seidell JC, Visser M, Deeg DJ, Dekker JM, et al. Vitamin D status and parathyroid hormone
levels in relation to blood pressure: a population-based study in older men and women. J Intern Med 2007
Jun;261(6):558-65.
(16) Reis JP, von MD, Kritz-Silverstein D, Wingard DL, Barrett-Connor E. Vitamin D, parathyroid hormone levels,
and the prevalence of metabolic syndrome in community-dwelling older adults. Diabetes Care 2007
Jun;30(6):1549-55.
(17) Norman AW, Bouillon R, Whiting SJ, Vieth R, Lips P. 13th Workshop consensus for vitamin D nutritional
guidelines. J Steroid Biochem Mol Biol 2007 Mar;103(3-5):204-5.
92
(18) Hermans MP, Levy JC, Morris RJ, Turner RC. Comparison of insulin sensitivity tests across a range of glucose
tolerance from normal to diabetes. Diabetologia 1999 Jun;42(6):678-87.
(19) Guillemant J, Le HT, Maria A, Allemandou A, Peres G, Guillemant S. Wintertime vitamin D deficiency in male
adolescents: effect on parathyroid function and response to vitamin D3 supplements. Osteoporos Int
2001;12(10):875-9.
(20) Hogstrom M, Nordstrom A, Nordstrom P. Relationship between vitamin D metabolites and bone mineral density
in young males: a cross-sectional and longitudinal study. Calcif Tissue Int 2006 Aug;79(2):95-101.
(21) Saquib N, von MD, Garland CF, Barrett-Connor E. Serum 25-hydroxyvitamin D, parathyroid hormone, and bone
mineral density in men: the Rancho Bernardo study. Osteoporos Int 2006 Dec;17(12):1734-41.
(22) Cauley JA, Parimi N, Ensrud KE, Bauer DC, Cawthon PM, Cummings SR, et al. Serum 25 HydroxyVitamin D and
the Risk of Hip and Non-spine Fractures in Older Men. J Bone Miner Res 2009 Sep 23.
(23) Kristinsson JO, Valdimarsson O, Sigurdsson G, Franzson L, Olafsson I, Steingrimsdottir L. Serum 25hydroxyvitamin D levels and bone mineral density in 16-20 years-old girls: lack of association. J Intern Med 1998
May;243(5):381-8.
(24) Outila TA, Karkkainen MU, Lamberg-Allardt CJ. Vitamin D status affects serum parathyroid hormone
concentrations during winter in female adolescents: associations with forearm bone mineral density. Am J Clin
Nutr 2001 Aug;74(2):206-10.
(25) Bischoff-Ferrari HA, Dietrich T, Orav EJ, wson-Hughes B. Positive association between 25-hydroxy vitamin D
levels and bone mineral density: a population-based study of younger and older adults. Am J Med 2004 May
1;116(9):634-9.
(26) Hannan MT, Litman HJ, Araujo AB, McLennan CE, McLean RR, McKinlay JB, et al. Serum 25-Hydroxyvitamin
D and Bone Mineral Density in a Racially and Ethnically Diverse Group of Men. J Clin Endocrinol Metab 2007
Nov 6.
(27) Bonjour JP, Theintz G, Law F, Slosman D, Rizzoli R. Peak bone mass. Osteoporos Int 1994;4 Suppl 1:7-13.
(28) Nordstrom P, Neovius M, Nordstrom A. Early and rapid bone mineral density loss of the proximal femur in men.
J Clin Endocrinol Metab 2007 May;92(5):1902-8.
(29) Wren TA, Kim PS, Janicka A, Sanchez M, Gilsanz V. Timing of peak bone mass: discrepancies between CT and
DXA. J Clin Endocrinol Metab 2007 Mar;92(3):938-41.
(30) Chapuy MC, Arlot ME, Duboeuf F, Brun J, Crouzet B, Arnaud S, et al. Vitamin D3 and calcium to prevent hip
fractures in the elderly women. N Engl J Med 1992 Dec 3;327(23):1637-42.
(31) Lenders CM, Feldman HA, Von SE, Merewood A, Sweeney C, Wilson DM, et al. Relation of body fat indexes to
vitamin D status and deficiency among obese adolescents. Am J Clin Nutr 2009 Sep;90(3):459-67.
(32) Lee DM, Rutter MK, O'Neill TW, Boonen S, Vanderschueren D, Bouillon R, et al. Vitamin D, parathyroid
hormone and the metabolic syndrome in middle-aged and older European men. Eur J Endocrinol 2009
Dec;161(6):947-54.
(33) Cheng S, Massaro JM, Fox CS, Larson MG, Keyes MJ, McCabe EL, et al. Adiposity, cardiometabolic risk, and
vitamin D status: the Framingham Heart Study. Diabetes 2010 Jan;59(1):242-8.
(34) Pittas AG, Harris SS, Stark PC, wson-Hughes B. The effects of calcium and vitamin D supplementation on blood
glucose and markers of inflammation in nondiabetic adults. Diabetes Care 2007 Apr;30(4):980-6.
(35) Forman JP, Giovannucci E, Holmes MD, Bischoff-Ferrari HA, Tworoger SS, Willett WC, et al. Plasma 25hydroxyvitamin D levels and risk of incident hypertension. Hypertension 2007 May;49(5):1063-9.
93
Table 1. Seasonal variation in serum 25-vitamin D in 700 young Danish men (cross-sectional study).
Data are shown as mean (sd).
Spring
Summer
Autumn
Winter
March-May
June-August
Sept.-Nov.
Dec.-Feb
91
99
199
311
700
59.3 (27.1)
60.6 (27.9)
75.0 (19.6)
70.4 (29.5)
64.9 (27.7)
Deficiency (n)
26.1 %
6.3 %
20.7 %
43.6 %
31.4 %
Insufficiency (n)
18.3 %
41.1 %
38.3 %
28.7 %
32.7 %
N
Serum 25OHD (nmol/L)
Chi2 test: p<0.001
Overall
94
Table 2. Vitamin D status and relation with lifestyle factors, body composition, bone mass and bone
markers. Data are shown as mean (sd) or median [quartiles]. * not normally distributed.
Deficiency
Insufficiency
Sufficiency
(n=220)
(n=240)
(n=240)
25.3 (2.8)
25.6 (2.8)
25.6 (2.9)
NS
Multivitamin supplementation (y/n)
8%
12 %
18 %
P=0.002
Participation in sports (y/n)
37 %
43 %
57 %
P<0.001
Alcohol (units)
9 [5-15]
8 [4-16]
8 [5-15]
NS
Present use of tobacco (y/n)
40 %
32 %
28 %
P=0.015
BMI (Kg/m2)
24.8 (4.1)
24.2 (2.9)
24.0 (3.2)
P=0.001
Waist/hip ratio
0.91 (0.07)
0.89 (0.05)
0.88 (0.05)
P<0.001
BFAT (Kg) *
15.3 [11.1-20.3]
13.4 [10.5-17.3]
12.9 [10.0-16.0]
P<0.001
ViFM (cm2) *
39.1 [28.2-51.6] (n=90)
32.6 [25.5-48.9] (n=133)
31.6 [24.6-41.5] (n=141)
P=0.021
ScFM (cm2) *
144.7 [93.3-215.8] (n=90)
136.8 [86.0-174.2] (n=133)
119.2 [74.1-160.4] (n=141)
P=0.004
LBM (Kg)
63.5 (7.5)
64.0 (7.3)
63.9 (6.7)
NS
Glucose (mmol/L)
5.2 (0.4)
5.2 (0.4)
5.2 (0.4)
NS
Insulin (pmol/L) *
37 [26-53]
33 [23-45]
31 [23-41]
P<0.001
HOMA (IR) *
1.2 [0.9-1.8]
1.1 [0.5-1.4]
1.0 [0.7-1.3]
P<0.001
Triglycerides (mmol/L)
1.4 (1.2)
1.1 (0.7)
1.1 (0.5)
P=0.003
LDL (mmol/L)
2.7 (0.9)
2.6 (0.8)
2.5 (0.7)
NS
HDL (mmol/L)
1.5 (0.3)
1.6 (0.4)
1.6 (0.4)
P=0.006
SBP (mmHG)
124 (13)
122 (12)
123 (12)
NS
DBP (mmHG)
75 (10)
74 (8)
74 (9)
NS
PTH (pmol/L)
2.5 [1.8-3.2]
2.2 [1.6-3.1]
2.0 [1.4-2.6]
P<0.001
Ionized calcium (mmol/L)
1.24 (0.03)
1.25 (0.03)
1.25 (0.03)
NS
BAP (U/L) *
26.5 [21.5-33.1]
24.4 [21.1-29.8]
24.5 [21.0-30.0]
P=0.003
1CTP (μg/L) *
4.7 [4.0-5.6]
4.5 [3.8-5.5]
5.0 [4.2-5.9]
P=0.007
Lumbar spine (g/cm2)
1.06 (0.12)
1.09 (0.13)
1.08 (0.12)
0.024
Femoral neck(g/cm2)
0.93 (0.13)
0.97 (0.14)
0.95 (0.13)
0.004
Total hip (g/cm2)
1.05 (0.14)
1.10 (0.14)
1.09 (0.14)
0.032
Age (years)
p-value
Lifestyle factors
Body composition
Insulin resistance, lipids and blood pressure
Calcium homeostasis and bone markers
Bone mineral density
95
Table 3. PTH and 25OHD: associations with body composition and glucose- and bone metabolism.
Analyses on all participants and stratified according to vitamin status.
All Participants
Sufficient
Unadjusted
Non-sufficient
Multiple regression analyses
25OHD
logPTH
25OHD
LogPTH
25OHD
logPTH
25OHD
logPTH
r
r
β (R2adj)
β
β (R2adj)
β
β (R2adj)
β
BMI (kg/m2) #
-0.10**
<0.01
-0.11** (0.02)
-0.05
0.07 (<0.01)
<0.01
-0.14** (0.02)
-0.08
WHR#
-0.22***
0.02
-0.17***(0.15)
-0.07
-0.05 (0.07)
0.05
-0.12** (0.15)
-0.13**
BFAT (kg)##
-0.18***
<0.01
-0.17*** (0.36)
-0.08*
-0.01 (0.27)
-0.05
-0.18*** (0.39)
-0.11**
ViFM (cm2) ##
-0.13*
-0.04
-0.14** (0.19)
-0.08
-0.04 (0.16)
-0.12
-0.17** (0.20)
-0.06
ScFM (cm2) ##
-0.14**
-0.05
-0.13* (0.27)
-0.09*
0.02 (0.23)
-0.09
-0.15* (0.28)
-0.10
LBM (kg) ###
0.03
0.03
0.10** (0.32)
0.06
0.02 (0.28)
0.10
0.14** (0.34)
0.06
Insulin resistance, lipids and blood pressure
Glucose (mmol/L)
-0.06
0.01
-0.04 (0.03)
<0.01
0.04 (<0.01)
-0.05
-0.06 (0.04)
0.02
Insulin (pmol/L)
-0.15***
-0.09*
-0.11** (0.26)
-0.12**
0.03 (0.13)
-0.06
-0.10* (0.30)
-0.14**
HOMA (IR)
-0.17***
-0.06
-0.11** (0.25)
-0.08*
0.03 (0.16)
-0.05
-0.11 (0.26)*
-0.10*
Triglycerides
-0.15***
-0.03
-0.11** (0.09)
-0.05
-0.03 (0.09)
-0.03
-0.06 (0.08)
-0.13**
LDL (mmol/L)
-0.12**
-0.06
-0.09* (0.11)
-0.09
-0.05 (0.11)
-0.09
-0.03 (0.09)
-0.08
HDL (mmol/L)
0.10*
-0.06
0.02 (0.17)
-0.08*
-0.02 (0.20)
-0.25***
0.06 (0.17)
<0.01
SBP (mmHG)
<0.01
-0.06
0.01 (0.06)
-0.07
0.08 (0.09)
-0.02
-0.06 (0.07)
-0.09
DBP (mmHG)
-0.03
-0.03
-0.03 (0.06)
-0.05
0.02 (0.09)
-0.06
-0.06 (0.05)
-0.05
Ionized calcium
<0.01
-0.27***
-0.08* (0.09)
-0.31***
0.14* (0.11)
-0.32***
-0.10 (0.10)*
-0.32***
BAP (U/L)
-0.10**
0.01
-0.09* (0.13)
0.02
-0.01 (0.16)
0.06
-0.12 (0.11)*
0.01
1CTP (μg/L)
0.10**
-0.13***
0.08* (0.32)
-0.07*
0.06 (0.23)
-0.06
<0.01 (0.36)
-0.08
BMDlumbar (g/cm2)
0.09*
0.03
0.06 (0.12)
0.05
-0.01 (0.09)
0.01
0.18*** (0.15)
0.08
BMDfemoral neck (g/cm2)
0.07
-0.02
0.05 (0.20)
0.02
-0.02 (0.17)
-0.02
0.19*** (0.23)
0.04
BMDtotal hip (g/cm2)
0.09*
<0.01
0.08* (.20)
0.02
-0.06 (0.15)
0.02
0.24*** (.26)
0.04
(mmol/L)
Calcium homeostasis
Bone mineral density
Adjustments for: age, alcohol, tobacco, sports, use of multivitamin supplements, and season and BMI (# not adjusted
for BMI ## model includes LBM, not BMI ### model includes BFAT, not BMI).
P-values: *p<0.05 **p <0.01 *** p<0.001
96
0
10
20
2
30
40
BAP (U/L)
PTH (ng/ml)
4
50
60
6
Fig.1. Association between 25OHD and PTH as well as BAP.
0
50
100
25OHD (nmol/L)
150
200
0
50
100
25OHD (nmol/L)
150
200
1.4
1.2
1
.8
.8
1
1.2
BMD: lumbar spine (g/cm2)
1.4
1.6
1.6
Fig 2. Association between 25OHD and BMD in the lumbar spine, total hip and femoral neck.
50
100
25OHD (nmol/L)
150
200
0
50
100
25OHD (nmol/L)
150
200
.6
.8
1
1.2
1.4
0
0
50
100
25OHD (nmol/L)
150
200
97
0
50
100
150
200
Fig 3. BMI in relation to 25OHD.
15
20
25
30
BMI (kg/m2)
35
40
98
22 Osteoporosis and Vertebral Fractures in Men aged 60-74 Years
M Frost1, Wraae K2, B Abrahamsen3, Hoiberg M1, C Hagen4, M Andersen1, and K Brixen1
1
Department of Endocrinology, Odense University Hospital, Odense, 2 Department of Internal
Medicine, Holstebro Hospital, 7500 Holstebro, Denmark, 3Department of Internal Medicine,
Gentofte Hospital, Copenhagen, 4 Department of Endocrinology, Bispebjerg Hospital, Copenhagen,
Denmark
99
Abstract:
Introduction and hypothesis: Limited information on the prevalence of osteoporosis and vertebral
fractures in men are available. Moreover, the choice of reference values for dual x-ray
absorptiometry (DXA) is debated. We evaluated the prevalence of osteoporosis and vertebral
deformities in a population-based sample of men.
Methods: Bone mineral density (BMD) was measured and vertebral deformities assessed using dual
energy x-ray absorptiometry (DXA) and vertebral fracture assessment (VFA), respectively, in a
random sample of 600 Danish men aged 60-74 years.
Results: The study population was comparable to the background population with regard to age,
body mass index and co-morbidity. Osteoporosis was diagnosed in less than 1% of the participants
at inclusion. Using Danish and NHANES III reference data, 10.2% and 11.5% of the study
population had osteoporosis, respectively. In all, 6.3% participants had at least one vertebral
fracture. BMD was significantly lower in participants with vertebral deformities, but only 24% of
these cases had osteoporosis.
Conclusion: Osteoporosis and vertebral fractures are prevalent in men aged 60-74 years. Although
the majority of deformities were present in individuals without osteoporosis, BMD was lower in
patients with vertebral fractures at all sites investigated. Male osteoporosis was markedly
underdiagnosed.
100
Introduction:
Osteoporosis is a leading cause of fractures in both men and women [1]. Although the incidence of
osteoporosis is lower in men than women after the age of 50 years, the lifetime risk of fractures in
men older than 50 years has been estimated to 21-24 % [2]. The worldwide distribution of fractures
is uneven as the prevalence of osteoporotic fractures is higher in Northern European and in
particular Scandinavian countries [3]. Since osteoporotic fractures confer significant disability [4],
decreased quality of life [5] and mortality [6], data on the epidemiology of osteoporosis are
important for the planning of preventive efforts. Population-based information on the prevalence of
osteoporosis in men is, however, limited.
Bone mineral density (BMD) and vertebral fractures are predictive of future vertebral as well as
non-spine fractures independently of other risk factors in men [7]. The lifetime risk of a clinical
vertebral fracture has been estimated to be 8.6% in Swedish men [2]. Because only 25-33% of
vertebral fractures are diagnosed clinically [8] and even sub-clinical vertebral fractures are
associated with significant additional fracture risk, morbidity [3] and mortality [9], this fracture type
is commonly a missed opportunity for treatment.
In a large prospective study on vertebral fractures in men, approximately 3/4 vertebral fractures
were caused by unknown factors or low-energy traumas, and 87% of the incident vertebral fractures
occurred in participants with T-scores >-2.5 [7]. This questions the use of DXA to detect
individuals at high risk of prevalent and future vertebral fractures.
Vertebral fracture assessment (VFA) is based on lateral images obtained from a bone densitometer,
and the method allows for evaluation of prevalent fractures without the radiation dose of
conventional radiography [10]. Although fractures observed by use of VFA and specific algorithmbased quantitative methods have been shown to predict future spine fractures in women
independently of age and BMI [11], information on the ability of VFA to detect incident fractures is
101
limited. Nevertheless, low radiation exposure, the possibility of combining DXA and an assessment
of vertebral fracture as well as the high sensitivity of VFA to detect moderate and severe fractures
[12] suggest that the method could prove useful as an addition to conventional bone mass
assessment by DXA. Different approaches to the identification of fractures on VFA are available.
The Genant semiquantitative (SQ) method is currently recommended by the International Society of
Clinical Densitometry due to ease with which it can be applied to clinical practice and a substantial
number of studies showing that the method is reliable and valid [13]. Other procedures have been
developed including an algorithm-based qualitative (ABQ) method that appears to identify the same
prevalence of radiographic fractures as the SQ method, whereas the latter identified more fractures
on VFA in women [14]. In a US study on more than 700 men aged 65+ years, a moderate
agreement between these two methods were found [15].
The aims of the present study were first to evaluate the prevalence of osteoporosis as defined by a
T-score of less than or equal to -2.5 at the total hip, femoral neck or lumbar spine in a populationbased sample of elderly men. Secondly, we aimed to determine the prevalence of vertebral fractures
using VFA. Thirdly, the impact of applying Danish reference values as opposed to NHANESIII
values was evaluated.
Subjects and methods:
Subjects
Participants in the study were recruited from a random sample of the background population issued
by the Danish Civil Registration System. First, 4,975 men aged 60-74 years and living in the county
of Funen received a questionnaire by mail. A second letter was issued to non-respondents and the
resulting response rate was 85 %. Second, an age-stratified random sample of 1,845 respondents
was asked to take part in further study procedures, which was accepted by 51%. Telephone
102
interviews were conducted in 47% of these respondents as 5 respondents died and another 56 were
prevented from attending the clinic due to current illnesses. In all, 697 of the 803 eligible
respondents agreed to take part in a clinical assessment, and inclusion of the study participants was
completed once 600 had attended clinical evaluation.
The study was approved by the local Ethics Committee and all participants received written and
oral information prior to giving written consent. The study was performed in accordance with the
Helsinki II declaration and registered in ClinicalTrials.gov as NCT00155961.
Body composition and bone mineral density
Body weight was measured in light clothing (SECA, Germany) to the nearest 0.1 Kg while height
was determined to the nearest 0.1 cm (Harpenden, Holtain, UK). BMI was calculated as weight
divided by height squared (Kg/m2). Bone mineral density (BMD) of the total hip, femoral neck and
lumbar spine were measured by use of dual x-ray absorptiometry (H4500, Hologic Inc., Waltham,
MA, USA). T-scores were calculated on the basis of a Danish male reference population (Mean
(SD): Spine: 1.073 (0.125)), femoral neck (0.948 (0.138)), total hip (1.078 (0.140)) [16] and a
combination of Hologic reference values for the spine (1.084 (0.111)) and NHANES III values for
the hip regions (Femoral neck: 0.930 (0.136), total hip 1.040 (0.144)).
Irrespective of reference material, osteoporosis was defined as a T-score at total hip, femoral neck
or the lumbar spine of equivalent to or less than -2.5.
Vertebral fracture assessment
The same densitometer was used for vertebral fracture assessment (VFA) on postero-anterior and
lateral (dual energy) images of the spine (T4-L4). As described by Rea et al. [17], the cranial and
caudal endplates were marked by 6 points at the anterior, middle and posterior part of the vertebrae.
The McCloskey algorithm [11] was used for the classification of fractures. In brief, a vertebrae
103
fracture was defined by 1) a 3 SD reduction in the ratio of vertebral heights compared to reference
values, either the anterior-posterior, middle-posterior or between adjacent posterior ratios or 2) a 3
SD reduction in the posterior height as predicted on the basis of the adjacent vertebrae [11]. The
algorithm classified the observed vertebrae as either fractured or not.
Kappa values for intra- and inter-observer agreement were 0.9 and 0.8, respectively.
Statistical methods
Results are presented as mean (SD) or median [25-75% percentiles] according to the observed
distribution. The study population was compared to the non-respondents using Pearson’s chi square
test or Student’s T-test as appropriate.
T-scores were calculated as the difference between the observed value and mean
reference value for young adult men divided by the standard deviation in young adult men.
Osteoporotic and non-osteoporotic as well as vertebral fracture and non-vertebral fracture
participants were compared by the use of chi square test, Student’s t-test or Mann-Whitney’s test, as
appropriate. The ability of osteoporosis to detect vertebral fractures was evaluated by calculation of
the area under the curve, and kappa statistics was used to evaluate the level of agreement between
the VFA interpretations.
All computations were performed using STATA v10 (STATA, College Station, TX,
USA), and p-values less than 5% were considered significant.
This study was supported by the World-Anti Doping Agency, Danish Ministry of
Culture and Institute of Clinical Research, University of Southern Denmark.
Results
104
The study population and those who were not included in the study were comparable with regard to
BMI and self-reported chronic illnesses, though smoking and pulmonary disease were observed less
frequently among respondents (24.4 % vs. 33.2%, p<0.001, and 6.1% vs. 9.6%, p=0.019,
respectively). Also, a higher proportion of the study population reported that they were living with a
partner (87.9% vs. 79.8%, p<0.001), participating in sports (15.6% vs. 8.1%, p<0.001) or had a
higher education (advanced studies: 32.1% vs. 22.4%, p<0.001) (Table 1).
In all, 2 participants reported a diagnosis of osteoporosis at inclusion.
BMD was assessed in 585 participants corresponding to 98% of the study population.
Using Danish (hip and spine) or NHANES III (with Hologic normative values used for the spine)
reference values resulted in a prevalence of osteoporosis in the study population of 10.2%, and
11.5%, respectively (Table 2). Substantial differences in the prevalence of osteoporosis were found
at all sites investigated, thus osteoporosis was substantially more common in the total hip and
femoral neck when Danish reference values were used (4.4% vs. 0.5% and 5.8% vs. 4.1%, both
p<0.01) whereas the Danish reference values resulted in a lower prevalence of osteoporosis in the
lumbar spine (4.6% vs. 8.0%, p<0.01) (Table 2).
VFA could be performed lege artis in 94% of the participants while evaluation of the
remaining scans was impossible due to severe arthritis and scoliosis in 1.2% and poor quality of the
scans in the remaining cases. None of the participants had signs of Scheuermann’s disease or
malignancy. Vertebral bodies were less frequently visible in the upper part of the spine. Thus, T6L4 and T8-L4 were visible in 48 % and 95 %, respectively (Table 4). In all, 35 (6.3%) participants
had at least one vertebral fracture comprising a total of 42 fractures consisting of 24 (62%) thoracic
and 16 (38%) lumbar fractures of which 13 fractures (31% of all fractures) were observed in 6
individuals with at least 2 fractures. The fractures were most prevalent in the thoraco-lumbar
junction and mid-thoracic spine (Table 4).
105
Including vertebral fractures as criteria for the diagnosis of osteoporosis increased the
prevalence of osteoporose to 14.8% and 15.7% depending on reference values used (Table 2). Two
(0.3%) of the study participants had been diagnosed with osteoporosis prior to study inclusion and
were treated with bisphosphonate.
Participants with osteoporosis as defined by Danish reference values were of similar
age as non-osteoporotic (Table 3). In men aged 60-64, 65-69 and 70+ years, the prevalence of
osteoporosis was 6.6%, 12.1% and 11.5%, respectively (p=0.06). BMI was lower in men with
osteoporosis (24.7 [23.3-27.6] vs. 27.5 [25.1-29.9] kg/m2, <0.001), whereas the intake of alcohol
was lower (6 [3-15] vs. 10 [6-19] units, p<0.01) and smoking more prevalent (35.5 % vs. 22.9%,
p<0.05) (Table 3).
Individuals with vertebral fractures on the basis of VFA had lower BMD at all sites
investigated compared to the remaining study population (BMDspine: 0.95 (0.88-1.10) vs. 1.05 (0.941.17) g/cm2, p<0.01), BMDtotal hip: 0.87 (0.83-0.98) vs. 0.96 (0.87-1.04) g/cm2, p<0.001, BMDneck:
0.70 (0.64-0.75) vs. 0.77 (0.69-0.84) g/cm2, p<0.001) (Table 3). In men aged 60-64, 65-69 and 70+
years, the prevalence of vertebral fractures was 2.8%, 9.1% and 5.9%, respectively (NS).
The proportion of individuals with a family history of osteoporosis or hip fracture or a
sedentary lifestyle was comparable in osteoporotic and non-osteoporotic as well as vertebral and
non-vertebral fracture participants (Table 3).
In all, data on both VFA and BMD were available in 92% of the participants. Twentyfour percent of those with a vertebral fracture had osteoporosis irrespective of reference values
used, whereas 1.5% had manifest osteoporosis defined as both a vertebral fracture and osteoporosis.
For patients with two or more vertebral fractures, 33% (2 out of 6) had BMD in the osteoporosis
range, which is significantly more than observed in individuals without fractures (p<0.05) (Table
3).
106
The diagnostic utility of BMD measurements for detection of coexisting vertebral
fractures was independent of the reference values used (area under the curve: 0.571 and 0.566,
respectively). The sensitivity of DXA using Danish and NHANESIII reference values and a T-score
of ≤ -2.5 to detect the presence of a concomitant vertebral fracture were the same (24%) whereas
the specificity was 91% and 90%, respectively. Positive and negative predictive values (PPV and
NPV) were also similar. PPV and NPV for BMD using Danish reference values were 14% and 95%
and the corresponding results for NHANESIII were 13% and 95%.
Discussion
This study provided for the first time observational data on the prevalence of osteoporosis as
determined by DXA on a population-based sample of elderly Danish men. On the basis of local
reference values [16], 10% of men aged 60-74 years had osteoporosis. Using NHANES III hip and
Hologic lumbar spine reference values provided slightly higher but clinically similar estimates of
osteoporosis in the study population. In contrast, there appeared to be substantial differences in the
prevalence of osteoporosis if only one region was evaluated, i.e. 4.4% and 0.5% of the participants
were osteoporotic if a Danish or NHANES III reference values for total hip were used, respectively
(Table 2).
The prevalence of osteoporosis in Danish men has previously been estimated to be 17.7% in those
aged more than 50 years on the basis of register-based data [18]. Our observations are not in
complete accordance with these results. This is probably due to the relatively limited range of age
covered in our study compared to the study population of men older than 50 years included in the
register-based study as well as the fact that fractures were not included in our study. Including
prevalent vertebral fractures in the definition would have increased the overall prevalence of
osteoporosis in the present study by almost 50%. In addition, BMD has been shown to decrease and
107
the prevalence of vertebral fractures as depicted on radiographs appears to increase with age in men
[7,19], therefore, the age of the participants is likely to have considerable impact on the prevalence
of both osteoporosis and vertebral fractures. Our data do not support such age-related differences,
but this may be due to the limited number of participants or the relative low prevalence of vertebral
fractures.
Six percent of our participants had a vertebral fracture on the basis of VFA, and most of the
vertebral fractures were observed in the thoracic spine, which is in accordance with previous studies
[20]. The prevalence of fractures in our study was lower than the 12 % observed in the 50-79 years
old men participating in the European Vertebral Osteoporosis Study (EVOS) [19], 30% observed in
Moroccan men of with a mean age of 64 years [21] and 32% in seen in US men of a mean age of 69
years [20]. In contrast, the result is higher compared to a Finnish study [22], showing a prevalence
of 2.8% in men aged 75+ years. These differences are likely explained by the method used for the
evaluation of fractures, differences in the diagnostic cut-off and in the recruitment of study
participants.
We cannot account for the mechanism causing the observed fractures, but two US studies reported
that the vast majority of vertebral fractures are due to low-energy traumas or of unknown origin
whereas only 18-26% of the fractures were caused by high energy trauma [7,23]. The lower levels
of BMD observed in study participants with a vertebral fracture indicates that these fractures are
associated with bone health and not merely due to high-energy traumas. Even if the latter was the
case, high-energy trauma fractures have been associated with increased risk of further fractures [7],
suggesting that the observed fractures confer significant risk of future fractures.
The prevalence of vertebral fractures was 10-13% in the US MrOS, which included community
dwelling men of significantly higher age than those in our study [24]. Information on incident
vertebral fractures is limited. In the US MrOS, the incidence was shown to be 2.2/1,000 person-
108
years. Osteoporosis was significantly more prevalent in those that experienced a vertebral fracture
(13% vs. 2%), but the vast majority of fractures occurred in individuals without osteoporosis [7].
Due to the cross-sectional design of our study, we were unable to interpret the sensitivity of
osteoporosis as defined by a T-score equal to or less than -2.5 to identify future vertebral fractures.
Using DXA for detection of concomitant vertebral fracture in the study population resulted in a
sensitivity of only 24%. In the Rotterdam study [25], the sensitivity of DXA, i.e. osteoporosis as
defined on the basis of DXA, to identify persons who would experience an incident non-vertebral
fracture was estimated to be 44 % and 21% in women and men, respectively. These results suggest
that T-score alone provides an inadequate evaluation of fracture risk.
VFA rather than X-ray was used to detect vertebral fractures in the present study. Compared to
radiographs, VFA appears to be less reliable in detecting grade 1 fractures in the upper thoracic
spine, however, conventional x-ray evaluations are also less precise in this region [26]. We used the
McCloskey algorithm for the evaluation of the VFA. This algorithm has been shown to predict
incident fractures as identified by radiographs in both men and women [27,28]. This algorithm has
also been applied to VFA and results in women showed that the McCloskey algorithm was able to
predict incident fractures independent of several factors including BMD [11]. Nevertheless,
although fractures are less common in men, we are not aware of any indications suggesting that the
method should not be appropriate in men.
Our results on VFA may have been influenced by problems of detecting grade I fractures, and it is
likely that radiographs would have allowed evaluation of a larger proportion of vertebrae. Although
most fractures occur between T7 and L3 [29], our estimate of the prevalence of vertebral fractures
is likely to be conservative. The present study is also weakened by the infrequency of osteoporosis
in the study population due to the age of the study population rather than a lack of external validity.
The limited frequency of the outcome of interest has probably weakened the interpretation of a
109
potential relation between age and osteoporosis or vertebral fractures. Adding to that, the results
may only be valid to Caucasian men aged 60-75 years.
Our study has a number of important strengths. First, the use of a population-based sample of men
limited the risk of selection bias. Second, acquisition of data by questionnaire in non-participants
allowed us to demonstrate that the study population was indeed comparable to the general
population of age- and sex-matched men. Our results, therefore, provide a reasonable estimate of
osteoporosis in elderly, Danish men. Thirdly, the lower levels of BMD observed in fracture patients
supported that these cases at least in part are related to poor bone mass and structure rather than to
reminiscences of prior high energy traumas.
While VFA may not detect grade I fractures or depict the upper part of the thoracic spine, data
support the use of this method in men with risk factors for fractures including a low T-score. The
International Society of Clinical Densitometry currently recommends the use of VFA in men with a
low T-score and one of several risk factors [13], nevertheless, our data suggest that VFA should be
used in all men evaluated with DXA.
In conclusion, in our population-based study on men aged 60-74 years, less than 1% reported
osteoporosis at inclusion but in fact a total of 10% had osteoporosis and 6% had at least one
vertebral. There was no overall substantial difference between results derived from using Danish
and NHANES-III reference values but there were significant differences with regard to the
prevalence of osteoporosis in specific regions, i.e. 4.4% had osteoporosis at the total hip using
Danish reference values while only 0.5% was osteoporotic on the basis of the NHANES values.
Moreover, our data demonstrate that VFA adds important information on bone status not captured
by T-scores derived from DXA.
110
Table 1. Comparison of the study population and individuals declining participation in the study.
Study population
Declined participation
Normal weight (BMI 18.5-24.9 kg/m2)
36.4%
34.5%
Overweight (BMI 25-29.9 kg/m2)
48.5%
49.7%
Obese (BMI >30 kg/m2)
14.9%
15.3%
Self reported healthy
47.6%
46.5%
Diabetes mellitus
6.5%
7.4%
Hypertension
22.1%
21.5%
Ischemic heart disease
12.6%
11.5%
Pulmonary disease
6.1%
9.6%*
Osteoporosis
0.2%
0.6%
Thyroid disease
0.5%
0.9%
Social and life style factors
Smoking
22.4%
33.2%***
Sports (1-3 hours/week)
15.6%
8.1%***
High school or technical studies
26.7%
24.2%**
Advanced studies
32.1%
22.4%***
Non-retired
20.9%
19.7%
Living with a partner
87.9%
79.8%***
*p<0.05 *** p<0.001
111
Table 2. Prevalence of osteoporosis according to reference values.
Measurements (n=585)
Danish reference
NHANES III
Lumbar spine (g/cm )
1.05 (0.18)
4.6 %
8.0 %* #
Femoral neck (g/cm2)
0.77 (0.12)
5.8 %
4.1 %*
Total hip (g/cm2)
0.95 (0.13)
4.4 %
0.5 %*
10.2 %
11.5 %*
Prevalence of osteoporosis defined as either low T-score, VFx or both
14.8%
15.8%*
Prevalence of both osteoporosis and VFx
1.5%
1.5%
2
Prevalence of osteoporosis
Prevalence of vertebral fractures: 6.3%
#
Hologic reference for lumbar spine used.
*
Prevalence significantly different from that derived from Danish reference values
112
Table 3. Comparison of participants with and without osteoporosis or vertebral fracture (VFx).
Danish reference values used for the diagnosis of osteoporosis.
Osteoporosis
No osteoporosis
VFx
No VFx
>1VFx
(n=61)
(n=524)
(n=35)
69 [65-71}
68 [64-71}
69 [67-71]
68 [64-72]
68 [67-68]
42.6%
38.4%
37.1%
39.6%
0%
24.6 [23.3-27.6]***
27.5 [25.1-29.9]
27.8 (25.5-30.7)
27.1 (24.7-29.6)
28.3 (26.1-29.8)
Lumbar spine (g/cm2)
0.80 [0.73-0.91]***
1.06 [0.96-1.18]
0.95 [(0.88-1.10]*
1.05 [0.94-1.17]
0.92 [0.83-1.05]
Femoral neck (g/cm2)
0.60 [0.57-0.64]***
077 [0.71-0.84]
0.70 [0.64-0.75]***
0.77 [0.69-0.84]
0.69 [0.59-0.73]*
Total hip (g/cm2)
0.74 [0.72-0.79]***
0.96 [0.90-1.05]
0.87 [0.83-0.98]***
0.96 [0.87-1.04]
0.84 [0.75-0.85]**
Osteoporosis
4.9%
6.7%
11.4%
6.3%
16.7%
Hip fracture
9.8%
9.0%
8.6%
9.2%
0%
36.3% *
22.9%
17.1%
24.1%
16.7%
Alcohol (units/week)
6 [3-15]**
10 [6-19]
8 [6-16]
10 [6-19]
12 (8-21)
Sedentary lifestyle¤¤¤
27.9%
24.4%
34.3%
23.7%
33.3%
(n=6)¤
Age and BMI
Age (years)
Proportion aged > 70 years
BMI (kg/cm2)
Bone mineral density
Family history of osteoporosis ¤¤
Life style factors
Smoker (yes/no)
*p<0.05, **p<0.01, *p<0.001
¤
Compared to the study population including those with one fracture.
¤¤
First degree relative with either osteoporosis or hip fracture
¤¤¤
Sedentary lifestyle defined as METS belonging to the lowest quartile
113
Table 4. Distribution of observed vertebrae and fractures in 558 men aged 60-74 years.
Vertebrae
Proportion of visible
Number of fractures
vertebrae
Proportion of vertebrae
with fractures
Th4
3
0.5 %
0
0
Th5
19
3.4%
0
0
Th6
270
48.4%
2
0.7%
Th7
475
85.1%
7
1.5%
Th8
531
95.2%
5
1.0%
Th9
544
97.5%
2
0.4%
Th10
550
98.6%
3
0.5%
Th11
551
98.7%
3
0.5%
Th12
554
99.3%
4
0.7%
L1
556
99.6%
9
1.6%
L2
557
99.8%
4
0.7%
L3
558
100%
1
0.2%
L4
556
100%
2
0.4%
114
Reference List
1
Genant HK, Cooper C, Poor G, Reid I, Ehrlich G, Kanis J, Nordin BE, Barrett-Connor E, Black D, Bonjour JP,
wson-Hughes B, Delmas PD, Dequeker J, Ragi ES, Gennari C, Johnell O, Johnston CC, Jr., Lau EM, Liberman
UA, Lindsay R, Martin TJ, Masri B, Mautalen CA, Meunier PJ, Khaltaev N, . Interim report and recommendations
of the World Health Organization Task-Force for Osteoporosis. Osteoporos Int 1999; 10: 259-64
2
Kanis JA, Johnell O, Oden A, Sembo I, Redlund-Johnell I, Dawson A, De Laet C, Jonsson B. Long-term risk of
osteoporotic fracture in Malmo. Osteoporos Int 2000; 11: 669-74
3
Johnell O , Kanis JA. An estimate of the worldwide prevalence and disability associated with osteoporotic
fractures. Osteoporos Int 2006; 17: 1726-33
4
Johnell O, Gullberg B, Kanis JA. The hospital burden of vertebral fracture in Europe: a study of national register
sources. Osteoporos Int 1997; 7: 138-44
5
Papaioannou A, Kennedy CC, Ioannidis G, Sawka A, Hopman WM, Pickard L, Brown JP, Josse RG, Kaiser S,
Anastassiades T, Goltzman D, Papadimitropoulos M, Tenenhouse A, Prior JC, Olszynski WP, Adachi JD. The
impact of incident fractures on health-related quality of life: 5 years of data from the Canadian Multicentre
Osteoporosis Study. Osteoporos Int 2009; 20: 703-14
6
Abrahamsen B, van ST, Ariely R, Olson M, Cooper C. Excess mortality following hip fracture: a systematic
epidemiological review. Osteoporos Int 2009; 20: 1633-50
7
Freitas SS, Barrett-Connor E, Ensrud KE, Fink HA, Bauer DC, Cawthon PM, Lambert LC, Orwoll ES. Rate and
circumstances of clinical vertebral fractures in older men. Osteoporos Int 2007;
8
Nevitt MC, Ettinger B, Black DM, Stone K, Jamal SA, Ensrud K, Segal M, Genant HK, Cummings SR. The
association of radiographically detected vertebral fractures with back pain and function: a prospective study. Ann
Intern Med 1998; 128: 793-800
9
Center JR, Nguyen TV, Schneider D, Sambrook PN, Eisman JA. Mortality after all major types of osteoporotic
fracture in men and women: an observational study. Lancet 1999; 353: 878-82
10 Blake GM, Rea JA, Fogelman I. Vertebral morphometry studies using dual-energy x-ray absorptiometry. Semin
Nucl Med 1997; 27: 276-90
11 McCloskey EV, Vasireddy S, Threlkeld J, Eastaugh J, Parry A, Bonnet N, Beneton M, Kanis JA, Charlesworth D.
Vertebral fracture assessment (VFA) with a densitometer predicts future fractures in elderly women unselected for
osteoporosis. J Bone Miner Res 2008; 23: 1561-68
12 Rea JA, Li J, Blake GM, Steiger P, Genant HK, Fogelman I. Visual assessment of vertebral deformity by X-ray
absorptiometry: a highly predictive method to exclude vertebral deformity. Osteoporos Int 2000; 11: 660-668
13 Schousboe JT, Vokes T, Broy SB, Ferrar L, McKiernan F, Roux C, Binkley N. Vertebral Fracture Assessment: the
2007 ISCD Official Positions. J Clin Densitom 2008; 11: 92-108
115
14 Ferrar L, Jiang G, Schousboe JT, DeBold CR, Eastell R. Algorithm-based qualitative and semiquantitative
identification of prevalent vertebral fracture: agreement between different readers, imaging modalities, and
diagnostic approaches. J Bone Miner Res 2008; 23: 417-24
15 Ferrar L, Jiang G, Cawthon PM, San VR, Fullman R, Lambert L, Cummings SR, Black DM, Orwoll E, BarrettConnor E, Ensrud K, Fink HA, Eastell R. Identification of vertebral fracture and non-osteoporotic short vertebral
height in men: the MrOS study. J Bone Miner Res 2007; 22: 1434-41
16 Hoiberg M, Nielsen TL, Wraae K, Abrahamsen B, Hagen C, Andersen M, Brixen K. Population-based reference
values for bone mineral density in young men. Osteoporos Int 2007
17 Rea JA, Steiger P, Blake GM, Fogelman I. Optimizing data acquisition and analysis of morphometric X-ray
absorptiometry. Osteoporos Int 1998; 8: 177-83
18 Vestergaard P, Rejnmark L, Mosekilde L. Osteoporosis is markedly underdiagnosed: a nationwide study from
Denmark. Osteoporos Int 2005; 16: 134-41
19 O'Neill TW, Felsenberg D, Varlow J, Cooper C, Kanis JA, Silman AJ. The prevalence of vertebral deformity in
european men and women: the European Vertebral Osteoporosis Study. J Bone Miner Res 1996; 11: 1010-1018
20 Vallarta-Ast N, Krueger D, Wrase C, Agrawal S, Binkley N. An evaluation of densitometric vertebral fracture
assessment in men. Osteoporos Int 2007; 18: 1405-10
21 El MA, Mounach A, Gassim S, Ghazi M. Vertebral fracture assessment in healthy men: prevalence and risk factors.
Bone 2008; 43: 544-48
22 Santavirta S, Konttinen YT, Heliovaara M, Knekt P, Luthje P, Aromaa A. Determinants of osteoporotic thoracic
vertebral fracture. Screening of 57,000 Finnish women and men. Acta Orthop Scand 1992; 63: 198-202
23 Cooper C, Atkinson EJ, O'fallon WM, Melton LJ, III. Incidence of clinically diagnosed vertebral fractures: a
population-based study in Rochester, Minnesota, 1985-1989. J Bone Miner Res 1992; 7: 221-27
24 Ferrar L, Jiang G, Cawthon PM, San VR, Fullman R, Lambert L, Cummings SR, Black DM, Orwoll E, BarrettConnor E, Ensrud K, Fink HA, Eastell R. Identification of vertebral fracture and non-osteoporotic short vertebral
height in men: the MrOS study. J Bone Miner Res 2007; 22: 1434-41
25 Schuit SC, van der KM, Weel AE, De Laet CE, Burger H, Seeman E, Hofman A, Uitterlinden AG, van Leeuwen
JP, Pols HA. Fracture incidence and association with bone mineral density in elderly men and women: the
Rotterdam Study. Bone 2004; 34: 195-202
26 Chapurlat RD, Duboeuf F, Marion-Audibert HO, Kalpakcioglu B, Mitlak BH, Delmas PD. Effectiveness of instant
vertebral assessment to detect prevalent vertebral fracture. Osteoporos Int 2006; 17: 1189-95
27 van der KM, De Laet CE, McCloskey EV, Johnell O, Kanis JA, Hofman A, Pols HA. Risk factors for incident
vertebral fractures in men and women: the Rotterdam Study. J Bone Miner Res 2004; 19: 1172-80
28 Ismail AA, Cockerill W, Cooper C, Finn JD, Abendroth K, Parisi G, Banzer D, Benevolenskaya LI, Bhalla AK,
Armas JB, Cannata JB, Delmas PD, Dequeker J, Dilsen G, Eastell R, Ershova O, Falch JA, Felsch B, Havelka S,
Hoszowski K, Jajic I, Kragl U, Johnell O, Lopez VA, Lorenc R, Lyritis G, Marchand F, Masaryk P, Matthis C,
Miazgowski T, Pols HA, Poor G, Rapado A, Raspe HH, Reid DM, Reisinger W, Janott J, Scheidt-Nave C, Stepan
116
J, Todd C, Weber K, Woolf AD, Ambrecht G, Gowin W, Felsenberg D, Lunt M, Kanis JA, Reeve J, Silman AJ,
O'Neill TW. Prevalent vertebral deformity predicts incident hip though not distal forearm fracture: results from the
European Prospective Osteoporosis Study
152. Osteoporos Int 2001; 12: 85-90
29 Hasserius R, Redlund-Johnell I, Mellstrom D, Johansson C, Nilsson BE, Johnell O. Vertebral deformation in urban
Swedish men and women: prevalence based on 797 subjects. Acta Orthop Scand 2001; 72: 273-78
117
23 Risk factors for fracture in elderly men. A population-based prospective study
Frost M1,2, Abrahamsen B2,3, Masud T2,4 and Brixen K1,2
1
Dept. of Endocrinology, Odense University Hospital, Odense, Denmark
2
Clinical Research Institute, University of Southern Denmark, Odense, Denmark
3
Dept. of Medicine F, Copenhagen University Hospital Gentofte, Hellerup, Denmark
4
Dept. of Geriatrics, Nottingham City Hospital, Nottingham, UK, and Department of Geriatrics,
Odense University Hospital, Odense, Denmark
118
Abstract
Introduction and hypothesis: Knowledge about risk factors for fracture in men is limited. The aim of
this study was to evaluate factors potentially associated with fracture risk in men.
Methods: A questionnaire enquiring about potential risk factors for fractures in men was posted to a
random sample of 9,314 men aged 60-74 years. A completed questionnaire was returned by 4,696
(50.4%). Follow-up on incident fractures over 5.4 years was performed using public registries.
Results: During the study, 203 individuals experienced a first clinical fracture, of which 85 patients
were considered osteoporotic (9 in humerus, 10 vertebral, 32 in the hip and 34 in the forearm). Cox
proportional hazard regression models were used to evaluate risk factors for any and osteoporotic
fractures. The following variables were found to be associated with increased risk of any fracture in
adjusted models family history of a hip fracture (HR; 95%CI: 1.56; 1.05-2.33), falls (2-4/year: 2.10;
1.35-3.27, >4/year: 2.46; 1.12-5.41, both compared to no falls), dizziness (2.36; 1.51-3.71), erectile
dysfunction (1.41; 1.06-1.87) and frequent urination (2.06; 1.26-3.39). Similarly, falls (2.36; 1.453.86), dizziness (2.83; 1.52-5.25), erectile dysfunction (2.01; 1.30-3.09) and pulmonary illness
(1.90; 1.03-3.53) were associated with increased risk of osteoporotic fractures in adjusted models.
Conclusion: These results underline the importance of assessment of dizziness, falls and those with
a family history of hip fracture. Frequent urination and erectile dysfunction were independently
associated with increased fracture risk. Although the mechanism of association is unknown, these
variables are likely to be indicators of frailty or hypogonadism.
119
Introduction
Fracture incidence is gender-specific, where fractures in young adults occur most frequently in men
and fractures in the elderly occur most frequently in women. The lifetime risk of fracture for
women above 50 years of age is approximately 50 % while that of men is around 20 % [1]. Of all
the osteoporotic fractures observed globally in the year 2000, 39 % were sustained by men [2]. In
addition, the consequences of fractures differ between the genders, as hip fractures have been
associated with twice the mortality rate in men as in women [3,4].
The causes of these differences in fracture incidence are presently not obvious. Intrinsic factors,
such as bone structure and strength and the abrupt decline in sex hormone levels observed at
menopause, as well as extrinsic issues, such as an increased risk of falls among women [5], may
explain some of the higher fracture incidence observed in older women.
A number of factors have been shown to increase the risk of a fracture in men. Fracture incidence
increases after the age of 50 years and in particular after the age of 75 years [6,7]. Low body mass
index (BMI) [8], smoking [9], walking disability [10,11], a family history of fracture [12] and a
fracture after the age of 50 years [11] have also been shown to contribute to fracture risk in men.
While it makes biological sense that many risk factors for fracture in women should also apply to
men, the number of prospective studies evaluating other potential risk factors for osteoporotic
fracture in men is limited. Cohort studies have linked type 2 diabetes (T2DM) [13], cardiovascular
disease [14] and hypertension [15] with osteoporotic fractures in men [13], suggesting that these
variables could prove useful in fracture prediction. Register-based studies have identified several
medications that were associated with fracture risk in men [16,17]. In addition to insights into the
pathophysiology of male osteoporosis, the risk conferred by these variables could be translated into
clinically useful predictor variables once confirmed in different populations. Recently, a WHO
collaborative centre has developed an assessment tool (FRAX) that predicts an individual’s 10-year
120
fracture risk on the basis of BMI, clinical risk factors and bone mineral density (BMD) in men and
women [18]. In addition, the Garvan fracture risk calculator [19] and the QFractureTM fracture
prediction algorithm have become available [13]. Even though several risk factors have been
established, osteoporosis remains under-diagnosed and inadequately treated in men [3,20].
The aim of the present study was to evaluate self-reported clinical risk factors for
fracture in a population-based cohort study in Danish men aged 60+ years. First, we wanted to
evaluate the importance of non-modifiable risk factors, such as age and family history of hip
fracture or osteoporosis. Second, we wanted to test a number of at least partly modifiable factors
potentially associated with fracture risk including i) life style and obesity, ii) mobility, iii) diabetes
and cardiovascular disease, iv) genitourinary illness, v) gastrointestinal illness and vi) certain
miscellaneous factors.
121
Subjects and methods
Subjects
The Study on Male Osteoporosis and Aging (SOMA) is an ongoing 10-year, prospective,
population-based study on aspects of aging in men. The study was performed as a single-centre
study of Danish men living in the Funen County. Recruitment of participants was based on the
Danish Civil Registration System (CRS). In Denmark, every citizen is assigned a unique 10-digit
identification number by birth. In the autumn of 2004, a questionnaire was posted to a random
sample comprising 9,314 men aged 60-74 years living in Funen County. By returning the
questionnaire and providing written consent, the responders accepted collection of data and a
register-based follow-up on incident diseases including fractures. Non-responders were sent a
reminder letter. In all, 4,939 questionnaires were returned with consent to participate; however, only
4,696 of these were completed, rendering the response rate 50.4 %.
The study was approved by the local ethics committee (reference number: 20030230) and the
Danish Data Protection Agency, and was listed at clinicaltrials.gov (NCT00463411).
Registry data
Using the individual’s CRS identification number, information on marital status, education,
employment and socioeconomic data as of 2004 were retrieved from the national demographic
databases of Statistics Denmark (ref.no. 703026). Data on all hospital contacts since 1977 and visits
to outpatient clinics since 1999 were retrieved from the Danish National Hospital Registry (DNHR).
Information on morbidity was retrieved from the DNHR, and the Charlson comorbidity index [21]
was calculated on the basis of diagnoses as classified in the International Classification of Disease
10th Revision (ICD-10) registered during the last 3 years prior to inclusion.
122
Local hospital discharge records were used for follow-up for vital status and incident fractures. All
first fractures after the completion of the questionnaire were recorded and categorized as either any
fracture or osteoporotic, the latter including forearm, humerus, clinical vertebral and hip fractures.
In order to evaluate the representativeness of the study cohort, their socioeconomic and morbidity
data were compared with the data for all Danish men belonging to the same age group in autumn
2004, totalling 339,885.
Questionnaires
The questionnaire covered 41 items including self-reported height and weight (continuous) as well
as loss of height and weight since the age of 25 years (categorical: yes/no). Further items asked
about use of tobacco (categorical: present, past and never smoker) and alcohol (categorical: 0, 1-7,
8-14, 15-21, >21 units or unknown). Physical activity items included any difficulties in walking
(categorical: yes/no), use of walking aids (categorical: yes/no), and falls during the last year
(categorical: times/last year: 0, 1, 2-4, 5-8, 9-12, >12, unknown). Participants were asked about
medical history, current disorders and symptoms from all organ systems (categorical: yes/no; those
responding ‘yes’ were asked to write the diagnosis or symptom(s)). Participants were specifically
asked about previous testicular disease (categorical: yes/no) and treatment with testosterone
(categorical: yes/no). In this study, hypogonadism was defined as bilateral orchiectomy in the
absence of prostate cancer or prescription of testosterone. Erectile dysfunction (ED) was evaluated
by the question “do you suffer from impotence” (categorical: yes/no). The variables ‘dizziness’ and
‘frequent urination’ were derived from self-reported symptoms. Family history items included hip
fracture after the age of 50 years (first-degree family) and familial disposition for osteoporosis
(first-degree family).
123
The questionnaires were scanned electronically. Items covering medical history and present
symptoms could be answered in writing. If the computer was unable to interpret a written response
and two staff members were unable to interpret the answer, the response was coded blank.
Statistics
Comparisons with the age-matched Danish male population with respect to register-based data on
socioeconomics and morbidity were made using Student’s t-test, Mann-Whitney test (median
Charlson index) or chi-square, as appropriate. Data on responders and non-responders were
assessed separately.
Cox proportional hazard regression models were used to establish the relative hazard and 95%
confidence interval for the initial fracture or primary osteoporotic fracture (clinical vertebral, hip,
forearm, humerus) using a time line from September 2004 to the first fracture, death or end of study
in January 2010.
Models including self-reported variables and register-based data were used. All variables used in
the models were present in at least 1 % of the participants. The first model included non-modifiable
variables including age in 3 strata (60-65, 65-70 and >70 years) and a family history of hip fracture
or osteoporosis. All subsequent models incorporated a number of potential risk factors as well as the
variables from the first model. In the second model, BMI in 5 strata (<18.5, 18.5-25, 25-30, 30-35
and >35 kg/m2), smoking status (dichotomous: present/previous or never smoker), alcohol intake
(none, 1-6, 7-14, 15-21, >21 units/week) and weight loss since the age of 25 years were applied.
The third model evaluated mobility by including use of arms for chair stand, walking disabilities,
dizziness and falls (0, 1, 2-4 or >4 falls per year). In sub-analysis, the effect of falls on fracture risk
was assessed after stratification according to age group or BMI. Type II diabetes (T2DM),
hypertension, ischemic heart disease (IHD), stroke, hyperlipideamia, atrial fibrillation, gout and
124
erectile dysfunction (ED) were all included in the fourth model. Enlarged prostate, hypogonadism
(including prescription of testosterone or bilateral orchiectomy but not prostate cancer) and frequent
urination were fitted in the fifth model. The sixth model included diarrhoea and peptic ulcer, while
the last model included rheumatoid arthritis, epilepsy and pulmonary disorders (COPD, asthma and
chronic bronchitis).
For each factor identified as having a significant association with fracture risk, population
attributable risk (PAR) was calculated as prevalence multiplied by the hazard ratio [22].
Calculations were done using STATA version 10 (StataCorp, College Station, TX, US).
Significance level was set at p<0.05 using two-sided tests.
Results
Comparison of responders, non-responders and general population
The 4,696 responders were slightly older than non-responders and the background population. The
fraction of responders who were married was significantly greater than among non-responders and
the background population, and significantly fewer were widowers (Table 1).
Personal net income was substantially higher in responders compared to non-responders and the
general population (Table 1). In addition, significantly more responders were on pre-retirement
pension, whereas fewer were on retirement or disability pension (Table 1).
Significantly more responders had a Charlson index of 0 compared to non-responders and the
background population (Table 1). The number of individuals with a Charlson index exceeding 6
was higher among non-responders than in the background population (Table 1).
Incident fractures
125
During a mean follow-up of 5.4 years (range 0-5.4 years), 203 participants experienced one or more
fractures (only the first fracture in each participant was considered). In all, 85 patients suffered
osteoporotic fractures (9 in humerus, 10 clinical vertebral, 32 in the hip and 34 in the forearm).
Model 1: Non-modifiable factors
Age was not significantly associated with risk of either an osteoporotic or any fracture. In univariate
analyses, family history of hip fracture or osteoporosis was significantly associated with fracture
risk. After adjustments for BMI and the other variables in the model, only family history of hip
fracture remained associated with the risk of any fracture (HR=1.56; 95%CI 1.05-2.33) (Table 2).
Model II: Life style factors and obesity
Based on self-reported body weight and height, mean BMI was 26.8 ± 4.3 kg/m2 (Table 2). Weight
loss since the age of 25 years was reported by 8.6 %, while 27.2 % were current smokers and 46.5%
were previous smokers (Table 1). Alcohol consumption exceeding 21 units per week was reported
by 10.2%.
Neither smoking nor alcohol consumption was significantly associated with fracture risk. Those
with a loss of weight after the age of 25 years had an increased risk of any and osteoporotic
fractures in the complete model including age (HR=1.62; 95%CI 1.05-2.49 and 2.17; 1.18-3.97,
respectively), whereas no association was seen with the other variables.
Model III: Mobility
Eight per cent of the participants reported always using chair arms when standing up from a chair,
13.5 % experienced some level of walking impairment, 4.6 % had dizziness and 19.6 %, 6.6 % and
1.5 % reported at least one, 2-4 or more than 4 falls during the previous year, respectively. When
126
compared to men who did not report a history of a fall, those who reported one fall were
significantly more likely to fracture (any: HR=1.56; 95%CI: 1.06-2.31 and osteoporotic: 2.59; 1.564.30, respectively). For any but not osteoporotic fractures, a higher prevalence of self-reported falls
was associated with increased hazard ratios (2-4 falls pr year: HR=2.18; 95%CI 1.41-3.37 and >4
falls/year: 2.46; 1.12-5.41) (Table 2). Reporting dizziness was associated with any and osteoporotic
fracture (HR=2.34; 95%CI 1.49-3.68 and 2.81; 1.48-5.32, respectively) (Tables 2 and 3). Neither
the use of chair arms for standing up or walking difficulty was significantly associated with fracture
risk (Tables 2 and 3).
In individuals with a BMI below 25 kg/m2, the risk of fracture was higher among those reporting at
least 4 falls (HR=6.6; 95%CI 2.6-16.5) compared to those with a single fall (HR=2.5; 95%CI 1.44.6): There was no such effect in men with a BMI exceeding 25 kg/m2 (Table 4). Stratifying
according to age group showed increasing risk of fractures among those older than 70 years with
multiple falls compared to those with only one fall in the year prior to the survey (HR=4.6; 95%CI
1.4-15.1 vs. 1.7; 0.8-3.7) (Table 4).
Model IV. Diabetes and cardiovascular disease
Men reporting hypertension had an increased risk of any fracture in the complete model including
adjustments for age and BMI compared to those without hypertension (HR=1.41; 95%CI 1.011.97). In the same model, men who reported erectile dysfunction (ED) had an increased risk of both
any and osteoporotic fractures compared to those not reporting ED (HR=1.48; 95%CI 1.12-1.96
and 2.18; 1.40-3.39, respectively): Neither of the other variables influenced fracture risk (Table 2).
Model V. Genitourinary condtions
127
Men reporting frequent urination were more likely to fracture (any: HR=2.57; 95%CI 1.58-4.18 and
osteoporotic: 2.80; 1.35-5.82, respectively). There were no incident fractures among participants
reporting hypogonadism.
Model VI. GI-tract related symptoms
Neither of the variables related to the gastrointestinal tract was associated with fracture risk.
Model VII. Miscellaneous variables
Men reporting COPD, asthma or chronic bronchitis were more likely to experience an osteoporotic
fracture compared to men without pulmonary disease (HR=2.16; 95%CI 1.17-3.98). Neither
rheumatoid arthritis nor epilepsy was associated with fracture risk (Table 2).
Complete model
Any fracture
In the model including all variables demonstrated in separate models to be associated with any
fracture risk, participants reporting family history of hip fracture had an increased fracture risk
(HR=1.68; 95%CI 1.14-2.49). Similarly, men with a history of one fall (HR=1.47; 95%CI 1.002.17), 2-4 falls (2.10; 1.37-3.22) or more than 4 falls (2.34; 1.08-5.07) in the past year or dizziness
(2.00; 1.28-3.14) were all more likely to fracture. Equally, both men with ED and those with
frequent urination had an increased risk of any fracture (HR=1.39; 95%CI 1.04-1.84 and 2.05; 1.253.36, respectively) (Table 2).
Osteoporotic fracture
Individuals reporting more than one fall in the previous year and those reporting dizziness had an
increased risk of fracture (HR=2.43; 95%CI 1.46-4.03 and 2.72; 1.45-5.10, respectively). Also, men
128
with ED (HR=2.04; 95%CI 1.31-3.18) were significantly more likely to sustain a fracture.
Participants with a pulmonary disorder (HR=1.95; 95%CI 1.05-3.62) but not men with weight loss
since the age 25 years (HR=1.68; 95%CI 0.93-3.06) had an increased fracture risk.
PAR
The PAR estimates ranged from 8.2 for more than 4 falls per year to 27.5 for 2-4 falls per year for
any fracture, and PAR for at least one fall in the preceding year was 64.4 for osteoporotic fractures
(Table 5). PAR estimates for dizziness were 22.6 and 38.4 for any and osteoporotic fractures,
respectively. The PAR estimates for frequent urination and pulmonary disease were 18.2 and 27.5
for any and osteoporotic fractures, respectively.
Discussion
In this prospective population-based study on fractures in elderly Danish men, self-reported erectile
dysfunction was found to be associated with an increased risk of fractures. To the best of our
knowledge, this is the first report on an association between ED and fracture risk in men. We also
found an association between frequent urination and an increased risk of any as well as osteoporotic
fractures. The study results further emphasized the importance of falls and dizziness as well as
family history of hip fracture and pulmonary disorders for risk of fractures.
Whether ED is sentinel to future fractures in men is tentative. BMD and ED were
previously not found to be associated with fracture risk in a small cross-sectional study of 75 men
aged over 50 years [23]. In our large cohort, we found ED to be associated with fracture risk after
adjustment for multiple factors potentially causing the dysfunction. ED probably captures effects
not accounted for in our study, including low serum testosterone and vascular insufficiency. Future
studies incorporating biochemical tests are needed to evaluate the association so that adjustment can
129
be made for sex hormone levels. Several studies have evaluated the importance of hypogonadism or
low levels of sex steroids on fracture risk [24-26]. Although medical history allowed us to evaluate
the effects of hypogonadism defined on the basis of prescription of testosterone or bilateral
orchiectomy (n=48), fracture risk appeared not to be increased in these cases; the number of
participants with these conditions was low, however.
Few studies have evaluated the association between self-reported frequent urination
and fracture risk in men. In a retrospective study, increased nocturnal micturition and nocturnal
urine output were associated with risk of hip fracture [27] whereas nocturia has been shown to be a
risk factor for incident hip fractures [28]. Nocturia could influence fracture risk in men through an
increased incidence of falls. Although frequent urination appeared to influence fracture risk
independently of falls in our study, we cannot exclude the possibility of an explicit effect of
nocturia on fracture risk.
Reporting a family history of hip fracture but not osteoporosis was found to be
associated with an increased risk of any fractures, which is in line with the meta-analysis by Kanis
et al. [29] showing that a parental history of a fracture was associated with an increased risk of an
osteoporotic fracture, whereas in a substantially larger UK study, a family history of osteoporosis
and future fractures in men were unrelated [13].
Our results emphasize the significance of falls in the context of fractures and
contribute new information on dizziness as a relevant predictor of fractures in men. These findings
were unaffected by variables likely to include information on balance and agility, such as walking
disabilities. Recently, falls were found to increase the risk of osteoporotic fractures in UK men [13].
It appears that fall prevention is essential to fracture prevention in men as in women.
Although dizziness probably predisposes to falls, our results may be confounded by
several factors. Physical incapacities causing falls or dizziness could have been captured by
130
adjusting for self-reported walking difficulties or problems with standing from a chair, but none of
these variables were related to fracture risk.
The association between pulmonary disorder and an increased risk of osteoporotic
fracture is in line with previous research. COPD, asthma and use of inhalation and oral steroids
were associated with bone loss in both spine and hip as well as an increased risk of vertebral and
non-vertebral fractures in a US male cohort study [30], while others found an increased fracture risk
in men with asthma or currently on corticosteroids [13]. In addition, Vestergaard et al. [31] found a
1.2-1.3 times higher risk of fractures in patients with a chronic pulmonary disorder. Together, these
results underline the importance of assessment of pulmonary disorders in fracture prevention in
men.
From a theoretical point of view and presuming a cause-effect relationship between
the risk factor and fracture occurrence, the elimination of several factors found to increase fracture
risk could substantially reduce the prevalence of fractures. We did not combine all the PAR results
in a single estimate as many of the factors may be related, but falls, dizziness and frequent urination
are all potentially modifiable. These results suggest a potential for reducing the burden of fractures
in men by addressing a number of prevalent issues, including falls.
Advantages of the present study were the random selection of the study population
and the possibility to evaluate the representativeness of the study population. Although the study
participants were not comparable to the age-matched male background population on all
parameters, the differences were small. Our results may have been affected by the study population
being more healthy and wealthy, however, and this is important when extrapolating the study results
to the general population. None of the participants were lost to follow-up and, due to the public
nature of the Danish health system; there was free access to health care for all those with incident
fracture.
131
Our study had some limitations. The sensitivity of the approach could be questioned
as our data did not identify an effect of age, BMI, smoking and alcohol on fracture risk contrary to
other reports [6-9,13,32,33], although we did find an adverse effect of weight loss since adulthood
on fracture risk. Only baseline information was available for fracture prediction in this study, and
changes in comorbidity are very likely. Recall and information bias are likely to have influenced
some of the data, such as the recollection of hip fractures in relatives. Alcohol intake, height and
weight are likely to be influenced by report bias and should be interpreted with caution. We used a
public register for random selection of our study population; however, selection bias cannot be
excluded. In addition, data on fractures were obtained from the public register and thus depend on
proper classification and regular update. As spinal X-rays and biochemical tests were not part of the
study, we could not identify incident vertebral fractures not coming to clinical attention or adjust for
the levels of sex hormones and other biochemical factors.
In conclusion, we found erectile dysfunction and self-reported frequent urination to be
associated with fracture risk in elderly men. The results of this prospective population-based study
underline the importance to fracture prevention of assessment of falls, dizziness and pulmonary
disorders as well as weight loss since adulthood. Finally, our data support a positive association
between family history of hip fracture and increased fracture risk.
Acknowledgements
The study received financial support from the Institute of Clinical Research, University of Southern
Denmark, the Novo Nordisk Foundation and the Velux Foundation.
Conflicts of interest: none
132
Reference List
1
van Staa TP, Dennison EM, Leufkens HG, Cooper C. Epidemiology of fractures in England and Wales. Bone 2001;
29: 517-22
2
Johnell O , Kanis JA. An estimate of the worldwide prevalence and disability associated with osteoporotic
fractures. Osteoporos Int 2006; 17: 1726-33
3
Kiebzak GM, Beinart GA, Perser K, Ambrose CG, Siff SJ, Heggeness MH. Undertreatment of osteoporosis in men
with hip fracture. Arch Intern Med 2002; 162: 2217-22
4
Abrahamsen B, van ST, Ariely R, Olson M, Cooper C. Excess mortality following hip fracture: a systematic
epidemiological review. Osteoporos Int 2009; 20: 1633-50
5
Stevens JA , Sogolow ED. Gender differences for non-fatal unintentional fall related injuries among older adults.
Inj Prev 2005; 11: 115-19
6
Singer BR, McLauchlan GJ, Robinson CM, Christie J. Epidemiology of fractures in 15,000 adults: the influence of
age and gender. J Bone Joint Surg Br 1998; 80: 243-48
7
van Staa TP, Dennison EM, Leufkens HG, Cooper C. Epidemiology of fractures in England and Wales. Bone 2001;
29: 517-22
8
De Laet C, Kanis JA, Oden A, Johanson H, Johnell O, Delmas P, Eisman JA, Kroger H, Fujiwara S, Garnero P,
McCloskey EV, Mellstrom D, Melton LJ, III, Meunier PJ, Pols HA, Reeve J, Silman A, Tenenhouse A. Body mass
index as a predictor of fracture risk: a meta-analysis. Osteoporos Int 2005; 16: 1330-1338
9
Hoidrup S, Prescott E, Sorensen TI, Gottschau A, Lauritzen JB, Schroll M, Gronbaek M. Tobacco smoking and
risk of hip fracture in men and women. Int J Epidemiol 2000; 29: 253-59
10 Michaelsson K, Olofsson H, Jensevik K, Larsson S, Mallmin H, Berglund L, Vessby B, Melhus H. Leisure
physical activity and the risk of fracture in men. PLoS Med 2007; 4: e199
11 Lewis CE, Ewing SK, Taylor BC, Shikany JM, Fink HA, Ensrud KE, Barrett-Connor E, Cummings SR, Orwoll E.
Predictors of non-spine fracture in elderly men: the MrOS study. J Bone Miner Res 2007; 22: 211-19
12 Kanis JA, Johansson H, Oden A, Johnell O, De Laet C, Eisman JA, McCloskey EV, Mellstrom D, Melton LJ, III,
Pols HA, Reeve J, Silman AJ, Tenenhouse A. A family history of fracture and fracture risk: a meta-analysis. Bone
2004; 35: 1029-37
13 Hippisley-Cox J , Coupland C. Predicting risk of osteoporotic fracture in men and women in England and Wales:
prospective derivation and validation of QFractureScores. BMJ 2009; 339: b4229
14 Sennerby U, Melhus H, Gedeborg R, Byberg L, Garmo H, Ahlbom A, Pedersen NL, Michaelsson K.
Cardiovascular diseases and risk of hip fracture. JAMA 2009; 302: 1666-73
15 Vestergaard P, Rejnmark L, Mosekilde L. Hypertension is a risk factor for fractures. Calcif Tissue Int 2009; 84:
103-11
133
16 Abrahamsen B , Brixen K. Mapping the prescriptiome to fractures in men--a national analysis of prescription
history and fracture risk. Osteoporos Int 2009; 20: 585-97
17 Richards JB, Papaioannou A, Adachi JD, Joseph L, Whitson HE, Prior JC, Goltzman D. Effect of selective
serotonin reuptake inhibitors on the risk of fracture. Arch Intern Med 2007; 167: 188-94
18 Kanis JA, Johnell O, Oden A, Johansson H, McCloskey E. FRAX and the assessment of fracture probability in men
and women from the UK. Osteoporos Int 2008; 19: 385-97
19 Nguyen ND, Frost SA, Center JR, Eisman JA, Nguyen TV. Development of prognostic nomograms for
individualizing 5-year and 10-year fracture risks. Osteoporos Int 2008;
20 Geusens P , Dinant G. Integrating a gender dimension into osteoporosis and fracture risk research. Gend Med 2007;
4 Suppl B: S147-S161
21 Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in
longitudinal studies: development and validation. J Chronic Dis 1987; 40: 373-83
22 Benichou J. A review of adjusted estimators of attributable risk. Stat Methods Med Res 2001; 10: 195-216
23 Keles I, Aydin G, Orkun S, Basar MM, Batislam E. Two clinical problems in elderly men: osteoporosis and erectile
dysfunction. Arch Androl 2005; 51: 177-84
24 Meier C, Nguyen TV, Handelsman DJ, Schindler C, Kushnir MM, Rockwood AL, Meikle AW, Center JR, Eisman
JA, Seibel MJ. Endogenous sex hormones and incident fracture risk in older men: the Dubbo Osteoporosis
Epidemiology Study. Arch Intern Med 2008; 168: 47-54
25 Leblanc ES, Nielson CM, Marshall LM, Lapidus JA, Barrett-Connor E, Ensrud KE, Hoffman AR, Laughlin G,
Ohlsson C, Orwoll ES. The effects of serum testosterone, estradiol, and sex hormone binding globulin levels on
fracture risk in older men. J Clin Endocrinol Metab 2009; 94: 3337-46
26 Amin S, Zhang Y, Felson DT, Sawin CT, Hannan MT, Wilson PW, Kiel DP. Estradiol, testosterone, and the risk
for hip fractures in elderly men from the Framingham Study. Am J Med 2006; 119: 426-33
27 Asplund R. Hip fractures, nocturia, and nocturnal polyuria in the elderly. Arch Gerontol Geriatr 2006; 43: 319-26
28 Temml C, Ponholzer A, Gutjahr G, Berger I, Marszalek M, Madersbacher S. Nocturia is an age-independent risk
factor for hip-fractures in men. Neurourol Urodyn 2009; 28: 949-52
29 Kanis JA, Johansson H, Oden A, Johnell O, De Laet C, Eisman JA, McCloskey EV, Mellstrom D, Melton LJ, III,
Pols HA, Reeve J, Silman AJ, Tenenhouse A. A family history of fracture and fracture risk: a meta-analysis. Bone
2004; 35: 1029-37
30 Dam TT, Harrison S, Fink HA, Ramsdell J, Barrett-Connor E. Bone mineral density and fractures in older men
with chronic obstructive pulmonary disease or asthma. Osteoporos Int 2009;
31 Vestergaard P, Rejnmark L, Mosekilde L. Fracture risk in patients with chronic lung diseases treated with
bronchodilator drugs and inhaled and oral corticosteroids. Chest 2007; 132: 1599-607
134
32 Olofsson H, Byberg L, Mohsen R, Melhus H, Lithell H, Michaelsson K. Smoking and the risk of fracture in older
men. J Bone Miner Res 2005; 20: 1208-15
33 Kanis JA, Johansson H, Johnell O, Oden A, De Laet C, Eisman JA, Pols H, Tenenhouse A. Alcohol intake as a risk
factor for fracture. Osteoporos Int 2005; 16: 737-42
Table 1. The SOMA study population. Comparison with non-responders and the complete Danish
population of men aged 60-74 years.
n
Age (y)
Income (£)
SOMA
4,939
65 (60-75)
17,544¤¤
Non-SOMA
4,373
66 (60-75)
13,434
Population
339,885
66 (60-75)#
15,479##
Civil status
Married
Widower
80.9%¤¤
6.3%¤¤
68.7 %
8.4 %
73.7%##
7.0 %#
Socioeconomics
Public old age pension
Pre-retirement pension
Public disability pension
31.2%¤¤
25.9 %¤¤
6.7 % ¤¤
39.2 %
23.2 %
11.1 %
37.1 %##
23.2 %#
8.5 %##
Charlson Index
0
1-2
3-4
4-5
6->
43.0 %¤¤
11.8 %
0.9 %
0.4 %
36.1 %¤¤
36.5 %
12.2 %
1.1 %
0.5 %
40.9 %
39.8 %##
10.9 %#
1.0 %
0.5%
39.1%##
Values as mean, median (min-max) or per cent
SOMA responders vs non-responders:
SOMA vs. general population:
¤
p<0.05, ¤¤ p<0.001
#
p<0.05, ##: p<0.001
135
Table 2. Predictors of any fracture.
Univariate
HR (95%CI)
Age: 60-65 years
Age: 65-69 years
Age: 70-74 years
Family history of HFx after 50 years
Osteoporosis in family
Fx/No.reporting
variable
80/1825
69/1552
54/54/1116
31/419
28/388
Univariate
Fracture
Model
HR (95%CI)
Comprehensive model with
significant findings from models 1-7
HR (95%CI)
Model 1. Unmodifiable risk factors
Model 1 including BMI
1
1.01 (0.73-1.40)
1.11 (0.79-1.57)
1.72 (1.17-2.52)*
1.65 (1.11-2.46)*
1
1.01 (0.73-1.40)
1.12 (0.79-1.58)
1.56 (1.05-2.33) *
1.47 (0.97-2.23)
BMI <18.5 kg/m2
BMI 18.5-25.0 kg/m2
BMI 25-30 kg/m2
BMI 30-35 kg/m2
BMI >35 kg/m2 kg/m2
Loss of weight since age 25
Alcohol: none
Alcohol: 1-21 units/week
Alcohol: >21 units/week
Present or past smoker
3/34
69/1491
105/2229
23/600
3/139
26/379
16/375
161/3587
21/457
158/3350
Falls: 0/year
Falls: 1/year
Falls: 2-4/year
Falls: >4/year
Dizziness
Walking disabilities
Arms for chair stand
133/3534
32/506
27/279
7/61
23/191
37/598
24/351
Type II diabetes
Hypertension
Ischemic heart disease
Stroke
Erectile dysfunction
Hyperlididaemia
Atrial fibrillation (chronic and paroxysmal)
Gout
13/276
50/839
16/360
2/50
98/1716
27/593
9/142
4/52
Enlarged prostate
Nucturia
Frequent urination
Hypogonadism
9/155
4/42
18/164
0/48
Peptic ulcer
9/216
Model 2. BMI, life style and obesity.
Model 2
adjusted for age
1
1
0.55 (0.17-1.75)
0.56 (0.18-1.81)
0.56 (0.18-1.76)
0.63 (0.20-2.02)
0.46 (0.14-1.52)
0.48 (0.14-1.62)
0.26 (0.05-1.28)
0.30 (0.06-1.52)
1.57 (1.04-2.37)*
1.62 (1.05-2.49)*
1
1
1.05 (0.63-1.76)
1.09 (0.65-1.83)
1.09 (0.57-2.10)
1.13 (0.58-2.17)
1.20 (0.86-1.67)
1.22 (0.87-1.72)
Model 3. Mobility.
Univariate
Model 3
adjusted for BMI and age
1
1
1.66 (1.13-2.44)
1.56 (1.06-2.31)*
2.44 (1.60-3.71)
2.18 (1.41-3.37)*
3.02 (1.41-6.45)
2.46 (1.12-5.41)*
2.79 (1.81-4.31)*
2.34 (1.49-3.68)*
1.46 (1.02-2.08)*
1.05 (0.68-1.60)
1.58 (1.03-2.42)*
1.15 (0.70-1.88)
Model 4. Diabetes and cardiovascular disease
Univariate
Model 4
adjusted for BMI and age
1.05 (0.60-1.84)
0.94 (0.53-1.67)
1.41 (1.02-1.93)*
1.41 (1.01-1.97)*
0.90 (0.59-1.65)
1.02 (0.60-1.74)
0.90 (0.22-3.61)
0.83 (0.20-3.34)
1.50 (1.14-1.97)*
1.48 (1.12-1.96)*
1.01 (0.67-1.51)
0.89 (0.58-1.38)
1.40 (0.72-2.73)
1.41 (0.72-2.76)
1.70 (0.63-4.58)
1.62 (0.60-4.37)
Model 5. Genitourinary.
Univariate
Model 5
adjusted for BMI and age
1.31 (0.67-2.55)
1.37 (0.70-2.69)
2.05 (0.76-5.51)
2.25 (0.83-6.07)
2.48 (1.53-4.02)*
2.57 (1.58-4.18)*
No fractures
Model 6. Gastrointestinal.
Univariate
Model 6
adjusted for BMI and age
0.93 (0.48-1.82)
0.94 (0.48-1.83)
Diarrhoea
4/63
1.40 (0.52-3.77)
1.68 (1.14-2.49)*
Univariate
Univariate
Epilepsy
Pulmonary disorder
Rheumatoid arthritis
3/67
20/311
4/68
1.01 (0.32-3.17)
1.45 (0.92-2.31)
1.30 (0.48-3.50)
1.40 (0.52-3.77)
Model 7. Miscellaneous.
Model 7
adjusted for BMI and age
1.00 (0.32-3.14)
1.44 (0.91-2.29)
1.33 (0.49-3.58)
1.35 (0.88-2.06)
1.47 (1.00-2.17)*
2.10 (1.37-3.22)*
2.34 (1.08-5.07)*
2.00 (1.28-3.14)*
1.34 (0.97-1.85)
1.39 (1.04-1.84)*
2.05 (1.25-3.36)*
136
Separate models are presented. The 2° column represents univariate and the 3° individual models
(Model 1), adjusted for age (Model 2-7) and BMI (Models 3-7). The 4° column shows a
comprehensive model including significant findings from every individual model.
* significant findings.
137
Table 3. Predictors of osteoporotic fracture.
Fracture
Model
Comprehensive model:
HR (95%CI)
significant findings from models 1-7
HR (95%CI)
Model 1. Unchangeable variables
Univariate
Model 1 including BMI
Univariate
HR (95%CI)
Age: 60-65 years
Age: 65-69
Age: 70-74
Family history of HFx after 50yr
Osteoporosis in family
Fx/No.reporting
variable
34/1871
26/1595
25/1145
14/436
12/404
BMI <18.5 kg/m2
BMI 18.5-25.0 kg/m2
BMI 25-30 kg/m2
BMI 30-35 kg/m2
BMI >35 kg/m2 kg/m2
Loss of weight since age 25
Alcohol: none
Alcohol: 1-21 units/week
Alcohol: >21 units/week
Present or past smoker
2/35
30/1530
45/2289
7/616
1/141
14/391
5/386
71/3677
7/471
63/3445
Falls: 0/year
Falls: 1/year
Falls: 2-4/year
Falls: >4/year
Dizziness
Walking disabilities
Arms for chair stand
52/3615
22/516
7/298
2/66
12/202
18/617
14/361
Type II diabetes
Hypertension
Ischemic heart disease
Stroke
Erectile dysfunction
Hyperlididaemia
Atrial fibrillation (chronic and paroxysmal)
Arthritis urica
6/283
18/871
9/367
2/50
49/1765
13/607
4/147
0/56
Enlarged prostate
6/158
Nucturia
Frequent urination
Hypogonadism
1/45
8/174
0/48
Peptic ulcer
Diarrhoea
2/223
3/64
1
1
0.90 (0.54-1.50)
0.89 (0.54-1.49)
1.21 (0.72-2.03)
1.20 (0.71-2.01)
1.87 (1.06-3.32)*
1.69 (0.93-3.07)
1.70 (0.92-3.12)
1.45 (0.77-2.74)
Model 2. Age, life style and obesity.
Univariate
Model 2
adjusted for age
1
1
0.35 (0.09-1.47)
0.34 (0.08-1.46)
0.35 (0.09-1.45)
0.39 (0.09-1.67)
0.20 (0.04-0.98)*
0.24 (0.05-1.18)
0.13 (0.01-1.40)
0.16 (0.01-1.78)
2.11 (1.19-3.75)*
2.17 (1.18-3.97)*
1
1
1.48 (0.60-3.67)
1.63 (0.65-4.08)
1.16 (0.37-3.65)
1.31 (0.41-4.16)
0.97 (0.60-1.58)
0.93 (0.57-1.52)
Model 3. Mobility.
Univariate
Model 3
adjusted for BMI and
age
1
1
2.90 (1.76-4.78)
2.59 (1.56-4.30)*
1.64 (0.75-3.62)
1.28 (0.57-2.89)
2.19 (0.53-8.99)
1.41 (0.33-6.02)
3.54 (1.92-6.52)*
2.81 (1.48-5.32)*
1.75 (1.04-2.95)*
1.15 (0.61-2.18)
2.76 (1.67-4.58)*
1.69 (0.84-3.39)
Model 4. Diabetes and cardiovascular disease
Univariate
Model 4
adjusted for BMI and
age
1.16 (0.51-2.67)
1.06 (0.45-2.48)
1.15 (0.68-1.94)
1.14 (0.66-1.96)
1.37 (0.69-2.74)
1.28 (0.61-2.66)
2.20 (0.54-8.93)
1.84 (0.45-7.55)
2.19 (1.43-3.37)*
2.18 (1.40-3.39)*
1.19 (0.66-2.14)
1.00 (0.53-1.90)
1.49 (0.55-4.06)
1.46 (0.53-4.01)
No fractures
Model 5. Genitourinary.
Univariate
Model 5
adjusted for BMI and
age
2.14 (0.94-4.91)
2.25 (0.97-5.21)
1.19 (0.17-8.57)
1.37 (0.19-9.86)
2.63 (1.27-5.45)*
2.80 (1.35-5.82)*
No fractures
Model 6. GI-tract related symptoms.
Univariate
Model 6
adjusted for BMI and
age
0.48 (0.12-1.96)
0.49 (0.12-2.00)
2.55 (0.80-8.06)
2.51 (0.79-7.96)
Model 7. Miscellaneous.
Univariate
Model 7
adjusted for BMI and
1.68 (0.93-3.06)
1
2.43 (1.46-4.03)*
1.22 (0.54-2.71)
1.47 (0.35-6.10)
2.72 (1.45-5.10)*
2.04 (1.31-3.18)*
1.72 (0.81-3.64)
138
Epilepsy
Pulmonary disorder
Rheumatoid arthritis
12/319
3/69
age
No fractures
2.19 (1.19-4.03)*
2.16 (1.17-3.98)*
2.39 (0.76-7.57)
2.51 (0.79-7.96)
1.95 (1.05-3.62)*
Separate models are presented. The 2° column represents univariate and the 3° individual models
(Model 1), adjusted for age (Model 2-7) and BMI (Models 3-7). The 4° column shows a
comprehensive model including significant findings from every individual model.
* significant findings
Table 4. Relation between number of falls and risk of any fracture stratified according to BMI or
age.
Age
60-65 years
65-69 years
70-75 years
BMI:
<25kg/m2
>=25kg/m2
One fall/year
2-4 falls/year
>=4 falls/year
1.5 (0.8-2.7)
1.9 (1.0-3.6)*
1.7 (0.8-3.6)
2.0 (1.0-4.1)*
2.8 (1.4-5.5)*
2.7 (1.2-6.0)*
2.2 (0.6-9.2)
2.6 (0.6-10.7)
4.6 (1.4-15.1)*
2.5 (1.4-4.6)*
1.3 (0.8-2.2)
2.4 (1.1-5.2)*
2.4 (1.5-4.0)*
6.6 (2.6-16.5)*
1.3 (0.3-5.2)
* significant findings
Table 5. Population attributable risk.
Any fracture
Family history of HFx after 50 years
Falls: 1/year
Falls: 2-4/year
Falls: >4/year
Dizziness
Frequent urination
Osteoporotic fracture
Falls: 1/year
Dizziness
Pulmonary disorder
Prevalence in fracture cases
Prevalence in
non-fracture cases
PAR
15.3%
16.2%
13.1%
3.5%
11.3%
8.9%
9.3%
11.6%
6.4%
1.4%
4.3%
3.7%
25.7
23.8
27.5
8.2
22.6
18.2
26.5%
14.1%
14.1%
11.5%
4.4%
6.9%
64.4
38.4
27.5
139
Appendix A. Variables used for calculation of the Charlson index. Information obtained from public
register.
Disease
Myocardial infarction
Congestive heart failure
Peripheral vascular disease
Cerebrovascular disease
Dementia
Chronic pulmonary disease
Rheumatologic disease
Peptic ulcer disease
Mild liver disease
Diabetes
Diabetes with chronic complications
Hemi- or paraplegia
Renal disease
Malignancy
Severe (and moderate) liver disease
Metastatic, solid tumours
AIDS
Codes
I21-2, I252
I43, I50, P290, I109, I110, I130, I132, I255, I420, I425, I426, I427, I428, I429
I70, I71, I731, I738, I739, I771, I790, I792, K551, K558, K559, Z959, Z958
G45, G46, I60-69, H340
F00-03, G30, F051, G311
J40-47, J60-67, I278-279, J684, J701, J703
M05-06, M32-34, M315, M351, M353, M360
K25-28
B18, K73, K74, K700-3, K709, K713-15, K717, K760, K762-64, K768-69, Z944
E100-1, E106, E108-11, E116, E118-21, E126, E128-31, E136, E138-41, E146, E148-9
E102- 5, E107, E112-15, E117, E122-25, E127, E132-35, E137, E142-45, E147
G81-82, G041, G114, G801-2, G830-34, G839
N18-19, I120, I131, N032-37, N052, N057, N250, Z490-2, Z940, Z992
C00-26, C30-34, C37-41, C43, C45-61, C63, C65-76, C81-85, C88, C90-97
I850, I859, I864, I982, K704, K711, K721, K729, K765-67
C77-80
B20-22, B24
Appendix B. Items in questionnaire.
What is your height without shoes
Have you lost height since age 25 years
What is your weight
Have you lost any weight since the age of 25 years
Do you participate in sports activities
Do you use fitness centre on a regular basis
Do you have any walking difficulties
- if yes, do you use walking aids
How many hours a day do you stand or walk
- none, less than one hour, 1-3 hours, 4 hours or more
How often have you experienced a fall in the last 12 months
- never, once, 2-4, 5-8, 9-12, more than 12 times
Do you use your arms for chair stand
- never, rarely, once in a while, often, always
Do you live in a home for elderly people
Have you ever had a bone mass assessment (bone scan / DXA)
Have sustained any fracture after the age of 50 years
Have any of your family members fractured a hip after the age of 50 years
- if yes, tic who: mother, father, sister, brother
Have any of your family members been diagnosed with osteoporosis
- if yes, tic: mother, father, sister, brother, son, daughter, other
Do you have impotence/erection problems
- if yes, when did it start
Have you been treated with male sex hormone (testosterone)
Have you had any testicular diseases (e.i. severe traumas, cysts, cancer, operations)
- if yes, please explain
Have you experienced any long term bed rest
Do you take any medication, supplements, herbal medicine or vitamins (tablets, sprays, drops etc.)
- if yes, state name of drug, strength and dose (if possible)
Have you ever been treated for osteoporosis
- if yes, state name of drug and start and cessation of treatment
Have you ever been treated with corticosteroids (i.e. prednisolone, Cortisol or dexamethasone)
- if yes, explain
140
Do you take calcium supplements
Do you take vitamin D supplements (i.e. tablets, cod liver oil)
Do you drink milk or eat breakfast with milk?
- if yes, state average daily use
Do you eat fermented milk products
- if yes, state average daily use
Do you eat cheese
- if yes, state average daily use
Have any parts of your stomach or bowels been removed
Do you suffer from epilepsy
Do you have dementia
Do you have any symptoms from / diseases related to
- nervous system (i.e. vertigo, paralyses, eye etc). If yes, state symptom or disease
- the heart (i.e. chest pain, atrial fibrillation, hypertension, myocardial infarction etc). If yes, state symptom or disease
- the lungs (i.e. bronchitis, asthma, dyspnea etc). If yes, state symptom or disease
- stomach and bowel (i.e. peptic ulcer, diarrhoea, constipation, blood in stools, inflammatory bowel disease etc). If yes,
state symptom or disease
- urinary tract (i.e. frequent urination, painful urination etc). If yes, state symptom or disease
- metabolic disease (i.e. diabetes, thyroid disease etc). If yes, state symptom or disease
-bone and joints (i.e. rheumatoid arthritis, arthrosis etc). If yes, state symptom or disease
- any other (i.e. hyperlipideamia, psychiatric disease, skin diseases etc). If yes, state symptom or disease
What kind of home do you live in (villa, apartment, semi-detached etc) – and do you own or rent your home
What characterises your smoking status: present smoker, previous smoker (year of cessation), never smoker
What is your average weekly intake of alcohol (Nil, 1-7 units, 8-14 units, 15-21 units, 22-28 units, >28 units)
Do you have any other information
141
24 Pattern of use of DXA scans in men. A cross-sectional, population-based study
Frost M1,2, Gudex C1, Rubin KH1,2, Brixen K1,2 and Abrahamsen B2,3
1
Dept. of Endocrinology, Odense University Hospital, Odense, Denmark
2
Clinical Research Institute, University of Southern Denmark, Odense, Denmark
3
Dept. of Medicine F, Copenhagen University Hospital Gentofte, Copenhagen, Denmark
142
Abstract:
Purpose: Clinical and socioeconomic factors associated with bone mass assessment (DXA) in men
have seldom been evaluated. This study aimed to evaluate factors associated with the use of DXA
in men.
Methods: Self-report information on prior DXA and osteoporosis risk factors were obtained from
the baseline data of a study investigating the health perspectives of men aged 60-75 years.
Socioeconomic and comorbidity data were retrieved from national registers. The FRAX algorithm
was used to calculate the absolute fracture risk. Regression analysis was used to identify factors
significantly associated with previous DXA scan.
Results: Of the 4,696 men returning questionnaires (50% response rate), 2.7% had prior DXA but
48% had at least one osteoporosis risk factor. Previous DXA was associated with oral
glucocorticoid treatment, secondary osteoporosis, rheumatoid arthritis, fracture after age 50, falls
within the previous year, smoking and higher age. Twenty-one per cent of men with prior DXA and
10% of men without prior DXA had greater than 20% risk of a major osteoporotic fracture within
the next 10 years. One-third of those with previous DXA had none of the FRAX osteoporosis risk
factors. When family history of osteoporosis and falls were included as risk factors, 18% with
previous DXA had no clinical risk factors for osteoporosis.
Conclusions: DXA was infrequent in this group of elderly men, despite the presence of risk factors
for osteoporosis. DXA was also used despite a low fracture risk. There is a need for improved
targeting of DXA scans for men at high risk.
143
Introduction
The risk of fracture in men over 50 years old is 20-27%, which is lower than the comparable risk in
women (50 %) (1;2). However, hip fractures in men are associated with twice the mortality rate of
women (3;4). While young men sustain more fractures due to high-energy trauma than women, the
majority of fractures in elderly men and women are caused by low-energy trauma. Some of these
fractures are preventable as osteoporosis treatment options are available for both sexes (5).
Furthermore, treatment with anti-osteoporosis medication has been found to be cost-effective in
men aged over 65 years with a self-reported clinical fracture (6).
The prevalence of osteoporosis in Danish men aged over 50 years is estimated to be 17.7% (7), but
only 1.3 % of men aged over 60 years of age used bisphosphonates (8). Bone mineral density
(BMD) is a strong predictor of fractures in men (9;10) and bone densitometry is the standard
method for diagnosing osteoporosis in both sexes. However, there appears to be a gender difference
with regard to the use of bone densitometry. In 2005 in Australia, bone densitometry was used four
times more often in women (11), even though the life-time risk of a fracture in women and men
over 50 years old is estimated to be 44% and 27%, respectively (12), and 39 % of all fractures occur
in men (13). This suggests that a lack of awareness either among patients, physicians or both has
led to osteoporosis being under-diagnosed and insufficiently treated in both men and women (3;1416).
Several factors are associated with an increased risk of fractures in men, and
development of the 10-year fracture risk algorithm FRAX allows identification of patients at high
risk of osteoporotic or hip fractures, with or without assessment by DXA scanning (17). We have
limited knowledge about the use of DXA in men, and factors associated with densitometric
evaluation in men are not yet clarified (15). Similarly, there is little information available about the
risk factors and the 10-year fracture risk in men who have previously undergone DXA scanning.
144
While a number of clinical risk factors are useful for the identification of men who should undergo
DXA, other factors including socioeconomic status could influence bone density testing in men.
Thus, Neuner et al. (18) found that women living in low income areas were less likely to undergo
bone density testing prior to (though not after) fracture, and Demeter et al. (19) showed that use of
BMD testing was positively associated with income in women.
The aim of the present study was firstly to evaluate the use of densitometry in a population of
elderly Danish men and, secondly, to identify factors associated with the use of DXA in men.
Subjects and methods
Subjects
The present study used baseline data from the Study of Osteoporosis and Male Aging (SOMA),
which is a population-based, prospective study on health in older Danish men. The SOMA study
population was recruited using a random sample of Danish men living in Funen County, identified
by the personal unique identification number that all Danish citizens acquire at birth.
In the autumn of 2004, an invitation to participate in the SOMA study and a questionnaire covering
aspects of men’s health were sent by post to 9,314 men aged 60-74 years. A reminder letter
including the questionnaire was sent to non-responders after one month.
The local Ethics Committee (reference number: 20030230) and the Danish Data Protection Agency
approved the study, and SOMA is listed at clinicaltrials.gov (NCT00463411).
Register-based information
Information on income and marital status as of October 2004 was retrieved from the public register
Statistics Denmark (ref.no. 703026). The Sociodemographic data on the Danish age- and sexmatched population (n=339,885) as of October 2004 were obtained from Statistics Denmark.
145
Questionnaire
The self-completed questionnaire comprised 41 items and included questions on diagnosis of
osteoporosis, past or present treatment for osteoporosis, family history of osteoporosis or hip
fractures, previous bone densitometric evaluation and previous fractures (including the
circumstances causing the fracture). Information was also requested on current weight and height
and reductions in either of these since the age of 25 years, physical mobility (walking impairment,
ability to stand up without the aid of chair arms, frequency of falls) and oral glucocorticoid use.
Respondents also provided information on living conditions and lifestyle factors such as use of
alcohol and tobacco. Excessive intake of alcohol was defined as consumption of more than 21 units
of alcohol per week. Tobacco use was classified binomially as smoker or non-smoker, the latter
including ex-smokers.
The questionnaire data were read into the computer electronically. Any readable or ambiguous
answer was assessed by two medical secretaries. If the answer was still uncertain, the item was left
blank.
Risk factors and FRAX
Absolute fracture risk can be calculated using the algorithm FRAX. Since the Danish version of
FRAX is currently under development, the Swedish version was used. Clinical risk factors for
fracture were defined as those incorporated in the FRAX scoring system, i.e. age, low body mass
index (BMI below 19 kg/m2), previous fracture, family history of hip fracture, present use of
tobacco, present use of oral glucocorticoids, rheumatoid arthritis (RA), secondary osteoporosis and
excessive use of alcohol (here: more than 3 units of alcohol per day). Secondary osteoporosis was
defined by self-report of thyrotoxicosis, hypogonadism (prescribed testosterone or bilateral
146
orchiectomy) or type I diabetes.
A computerized call was made to the FRAX website in order to calculate the respondents’
individual 10-year risk of major osteoporotic and hip fractures.
Statistics
Data are reported as median (quartiles) or percentages of the study population. The participants
previously evaluated by DXA were compared to the remaining study population using MannWhitney tests (age, BMI, annual income) or Chi-square tests (all other variables). Multivariate
logistic regressions were used to evaluate factors associated with previous bone densitometry. The
independent variables in the initial regression analysis were the clinical risk factors from the FRAX
algorithm. Falls and family history of osteoporosis were added in a second regression analysis, and
income and marital status were added in the third regression analysis. All significant variables were
included in a fourth model. Likelihood-ratio tests were used to test model fit.
A p-value of less than 0.05 was considered significant and all calculations were performed using
STATA v.10 (Stata Corp, College Station, Texas, USA).
Results
Of the 4,939 questionnaires returned, 4,696 were complete, giving a response rate of 50.4%. A total
of 126 (2.7%) men reported previous bone densitometric evaluation and 0.6% reported current
treatment for osteoporosis.
Compared to non-responders, the study participants were of a similar age (both groups median age
65 [60-75] years) but had significantly higher income (median 190,096 DKK; quartiles 135,175291,059 DKK vs. 162,558 DKK; quartiles 120,006-261,835 DKK; p<0,001), were less likely to be
on retirement pension (31.2% vs. 39.2% of non-responders, p<0.01), more often married (80.9% vs.
147
68.7% of non-responders, p<0.01) and had less comorbidity (43% had a Charlson index score of 0
compared to 36.5% of non-responders, p<0.01).
Compared to the Danish age- and sex-matched population, the study participants had higher annual
income (median 190,096 DKK vs. median 168,806 DKK; quartiles 122,244-275,576 DKK;
p<0.001), more often on retirement pension (31.2% vs. 27.1%, p<0.01), more often married (80.9%
vs. 73.7%,of the Danish population p<0.01) and had less comorbidity (43% had a Charlson index
score of 0 compared to 39.8% of the Danish population, p<0.01).
Socioeconomics, physical characteristics and life style
As seen in Table 1, study participants with previous DXA were significantly older, more likely to
be divorced or live in a rented home and had a lower annual income compared to participants
without prior DXA.
Participants with previous DXA had lower height and weight than those without prior DXA, and
more often reported height or weight loss after the age 25 years. DXA was used similarly among
men with a low BMI (Table 1).
Participants with previous DXA were more likely to be current smokers. They were also more
likely to report at least one fall with in the preceding year, to have walking impairment or to use
chair arms for support when standing up (Table 1).
Family history and clinical characteristics
Participants with previous DXA were more likely to have a family history of osteoporosis, but not
hip fracture (Table 2). They were also more likely to report any fracture after the age of 50 years,
hip fracture, spine fracture or wrist fracture and to have a diagnosis of osteoporosis (Table 2).
148
Regarding comorbidity, participants with previous DXA were more likely to have rheumatoid
arthritis, hypogonadism, thyrotoxicosis, inflammatory bowel disease, and chronic obstructive
pulmonary disease but not asthma or chronic bronchitis (Table 2). They were also more likely to be
treated with anti-osteoporosis medication or oral glucocorticoids, either currently or in the past
(Table 2).
FRAX risk factors and scores
In the study population as a whole, DXA scanning had been performed in 1.6% of those with no
FRAX risk factors and 3.0%, 5.9% and 9.9% of those with 1, 2 or 3+ risk factors, respectively. Of
the 126 participants reporting a previous DXA, 31.8% had no clinical risk factors and 39.7% had
only one risk factor (Table 3). When family history of osteoporosis and tendency to fall (i.e. more
than one fall in past year) where included as risk factors, only 17.5% of the participants with prior
DXA scanning had no risk factor, and 23.8% had at least three risk factors (Table 3).
Participants with previous DXA had significantly higher median FRAX risk scores for major
osteoporotic fractures and hip fractures than participants without DXA and were also more likely to
have a greater risk of major fracture within the next 10 years (Table 3).
Predictors of previous DXA
Multivariate logistic regression analysis showed that oral glucocorticoid treatment (OR 9.05;
95%CI 4.35-18.8), secondary osteoporosis (3.84; 1.79-8.23), rheumatoid arthritis (2.56; 1.07-6.15),
fracture after 50 years of age (2.42; 1.55-3.77), falls (2.01; 1.36-2.96), use of tobacco (1.62; 1.112.37) and age (1.05; 1.00-1.10) were associated with prior bone mass assessment. The models
explained a minor part of the use of bone mass assessment (R2: 0.07-0.09). Adding falls and a
family history of osteoporosis significantly improved the second model (p<0.001) whereas
149
inclusion of the variables used in model 3 did not significantly improve the model.
Discussion
The study results show that only 2.7% of the men in this study aged 60-75 years reported previous
bone mass assessment by DXA scanning, despite a high prevalence of risk factors for osteoporotic
fractures. Thus approximately one-third of the participants had never been scanned, despite having
greater than 15% risk of a major osteoporotic fracture within the next 10 years. Secondly,
approximately one-third of the men who had previously undergone DXA scanning appeared not to
have any of the risk factors used in the FRAX algorithm.
The low rate of DXA scanning among older men in this study is at odds with an estimated
osteoporosis prevalence of 17.7% among Danish men aged over 50 years, and is surprising as both
DXA scans and osteoporosis-specific treatments are paid for or reimbursed by the Danish health
system. The study population differed slightly from the Danish population in terms of comorbidity,
income and marital status. Although these differences may influence the study, in a clinical
perspective the results are likely to be generalizable to the Danish age- and-sex matched population
as a whole. Only 0.6% of the men in the present study reported previous or current treatment for
osteoporosis; this is similar to the findings of register-based studies in which 0.3% and 1.3 % of
Danish men aged 50 years and over were receiving osteoporosis-specific treatment in 2004 and
2006, respectively (7;8). These results suggest that there needs to be more widespread treatment of
osteoporosis in men.
Both patients and the general public in Denmark lack specific knowledge about important risk
factors for osteoporosis (21). Previous studies suggest that elderly men have little knowledge about
male risk factors for osteoporosis (22) and that men, in contrast to women, do not recognize
osteoporosis as a serious disease that they are susceptible to (23). Male patients may be unaware of
150
the importance of DXA or unwilling to have a DXA scan. Equally, limited awareness of the
significance of bone scans among both general practitioners and specialists or even reluctance to
refer patients for assessment may explain the low use of DXA scanning and inadequate referral
rates. Further studies are needed to clarify the separate effects of these factors on the use of DXA.
Two separate Australian studies found that 76% of male fracture patients aged 50 years and over
remained untreated for osteoporosis despite 27% of these 87 untreated men having prior low trauma
fracture (24), and only 25% of the patients who contacted their general practitioner after a fracture
were recommended treatment for osteoporosis (25). Collectively, these studies indicate that both
patients and medical staff would benefit from an increased awareness and understanding of
osteoporosis in men, including the role of risk factors.
Targeting patients at risk of fracture is required if the incidence of fractures is to be reduced.
However, even among participants in our study who had widely recognized risk factors for
osteoporosis (such as oral glucocorticoid use and hip fracture), the use of DXA was very limited.
Furthermore, virtually none of the study participants who reported diagnoses that were probably
made by internal medicine specialists (such as hypogonadism and rheumatoid arthritis) had been
assessed by DXA. In their review of the literature on osteoporosis screening guidelines and patterns
of BMD testing, Morris et al. (15) found a weighted average screening rate of 8% for post-fracture
patients and 9% for glucocorticoid-treated patients. These rates are comparable to those of the
present study for men, and from a previous study among women (26).
We found a higher likelihood of bone mass evaluation as the number of clinical risk factors or the
FRAX score for major osteoporotic or hip fracture increased. Despite this, DXA had been
performed in only 10% of men with at least three FRAX risk factors, and 10% of men not assessed
by DXA had a 10-year major osteoporotic fracture risk of at least 20%. In comparison, 36% of
Danish women with three or more risk factors reported a previous DXA (26). It is also of concern
151
that 32% of the bone assessments in the present study had been performed in men who had no
apparent clinical risk factor as defined by the FRAX algorithm. The results thus suggest an
inappropriate pattern of referral of Danish men for DXA. The Swedish FRAX model was used in
this study, but it is likely to be similar to a Danish model. It should be noted, however, that the
FRAX algorithm has been validated primarily for use among women rather than men. Among
women, it was found that substantially fewer (10%) without any risk factors were evaluated by
DXA (26), suggesting a gender difference in DXA referral patterns.
A previous study found no association between age or co-morbidity and BMD testing (15). In the
present study, however, secondary osteoporosis and rheumatoid arthritis were associated with an
increased likelihood of a bone assessment.
Limitations of the study
Information bias may affect our results. Although prior fractures may be recalled with some
precision, other factors including falls are likely to be influenced by recall bias. With a response rate
of 50% we cannot exclude the possibility that the responders were a biased sample, towards
individuals with a particular interest in osteoporosis due to own risk factors or diagnosis of
osteoporosis. In addition, reduced renal function, low compliance with osteoporosis medication and
other factors that may influence the pattern of referral to DXA were not assessed in this study.
Furthermore, the public’s awareness of osteoporosis in men may have improved since the data
collection for this study, which may have led to greater use of DXA among individuals with risk
factors or previous fracture. Similarly, physicians may have become more aware of the need for
DXA referral of patients at high risk of fracture including those with a previous fragility fracture.
It has become more common for Australian men to have DXA scanning, and similar changes cannot
be excluded in Denmark (11). The availability of DXA scanners also differs between countries with
152
significantly more scanners available in other Nordic countries and fewer in the UK, Spain and the
Netherlands in 2003 compared to Denmark (27). The pattern of use of DXA may be different in
countries with a higher or lower availability of scanners. Another limitation of our study is the low
number of participants with prior DXA scan; this decreases the power to detect predictive factors.
Finally, we cannot exclude that some patients were referred for DXA following the finding of
reduced bone mineralization on X-rays taken for other indications. Such patients may have no
clinical risk factors for osteoporosis. In addition, we do not know whether referral for DXA was
initiated by the study participant or a treating physician.
The results of the current study rely on participants’ self-report of treatment for osteoporosis and
comorbidity and are thus open to bias due to underreporting. Any underreporting would mean that
the number of participants likely to benefit from referral to DXA would be even higher. The level of
agreement between questionnaire self-report and medical records varies according to the health
condition in question; agreement with respect to cardiovascular disease and diabetes appears to be
satisfactory whereas there can be a significant discrepancy for diseases of the musculoskeletal
system (28;29). In a study involving 120 individuals, more fractures were reported by the patients
than were noted in their medical records; however, the participants were a convenience sample and
had taken part in a general health screening program (30).
Paganini-Hill et al. (31) found a high recall accuracy of hip fractures, where the proportion of falsepositive responses was less than 10%. The recall accuracy of factors such as family history of
osteoporosis or hip fractures remains to be clarified. Unexpectedly, we found no effect of
socioeconomic status on the likelihood of a prior DXA scan. Participants who had a DXA scan
were more likely to be divorced or living in a rented home. Although annual income was lower in
those reporting a prior DXA scan, this relationship was not significant after adjustment for age and
comorbidity.
153
Few studies have examined the association between socioeconomic status and bone density in men.
In elderly Australian men, spine BMD was highest in men with either the highest or lowest
socioeconomic status, but there were no differences in BMD at the femoral neck (32). While
income appears not to be associated with fracture risk, living with a partner may reduce the risk of
fracture (33;34) .
Although the men in the study sample had generally higher income and lower comorbidity than
their counterparts in the general Danish population, the results are likely to be valid for the Danish
population of elderly men. The use of DXA is less likely to have been influenced by socioeconomic
status, as DXA scans are provided free of charge in Denmark and osteoporosis treatments are
subsidized by the state. Furthermore, the risk of recall bias was reduced through the use of registerbased information on socioeconomics and comorbidity.
In conclusion, we found that the use of DXA bone assessment was lower than would be expected in
elderly Danish men in comparison to the prevalence of known risk factors. The likelihood of a
DXA test was, however, higher in men with several clinical risk factors and a high 10-year fracture
risk, and there was only a limited effect from socioeconomic factors. Targeting DXA scans towards
individuals at high risk needs to be improved so as to avoid inappropriate resource use. The
implementation of a computerized risk algorithm such as FRAX may advance the identification of
men most likely to benefit from bone assessment. It appears that current referral rates that rely on
awareness and recognition of a number of unquantified, albeit well-described and well-known risk
factors are too low.
Acknowledgements
The study received financial support from the Novo Nordisk Foundation and the Velux Foundation.
154
Table 1. Socioeconomic, body composition and lifestyle characteristics of study population
according to DXA scanning status.
Data are absolute numbers (percentages) or medians (quartiles).
Previous DXA
Yes
n= 126
No
n= 4570
68.2 (63.5-71.2)
66.2 (63.1-69.9)*
158,680 (115,671-249,656) *|*
76.2%
12.7%*
8.7%
5.6%
45.7%***
20.5%
14.2%***
191,636 (135,179-292,594)
81.1%
7.2%
6.5%
3.5%
30.8%
26.1%
6.2%
63.5%
23.0%**
1.6%
68.8%
14.0%
1.6%
77 (70-86)***
175 (170-180)*
25.3 (22.9-28.0)**
2.4%
19.1% ***
50.0% ***
82 (74-90)
176 (172-180)
26.3 (24.3-28.7)
1.0%
8.3%
24.9%
11.1%
42.9%
37.3%*
10.2%
46.6%
26.9%
37.3%***
33.3%***
16.7%***
21.2%
13.0%
7.8%
Age (years)
Socioeconomics
Annual income 1 (DKK)
Married
Divorced
Widower
Unmarried
Retirement pension
Early retirement pension
Disability pension
Living conditions
Owner of home
Rented home
Care homes
Body composition
Weight (kg)
Height (cm)
Body Mass Index (BMI, kg/m)
Low BMI (<19kg/m2)
Weight loss since aged 25 years
Height loss since aged 25 years
Life style factors
Alcohol > 21 units/week
Previous smoker
Current smoker
Mobility
Falls (at least 1/year)
Walking impairment
Always using arms for chair stand
1
includes pensions.
*p<0.05, **p<0.01, ***p<0.001
155
Table 2. Family history and clinical characteristics of study population according to DXA scanning
status.
Data are percentages.
%
Family history of hip fracture
Family history of osteoporosis
Fractures since the age of 50 years
Any fracture
Hip fracture
Spine fracture
Wrist fracture
Diagnosis of osteoporosis
Co-morbidity
Chronic bronchitis
Asthma
Chronic obstructive pulmonary disease
Inflammatory bowel disease
Thyrotoxicosis
Hypogonadism
Rheumatoid arthritis
Type 1 Diabetes
Type 2 Diabetes
Treatment with oral glucocorticoids
Current treatment
Previous treatment
Treatment for osteoporosis
Current or previous treated
*p<0.05, **p<0.01, ***p<0.001
Previous DXA
Yes
No
(n=126)
(n=4570)
10.3%
9.6%
15.1%*
8.7%
23.8%***
5.6%***
7.9%%***
4.8% %*
15.9%***
9.4%
0.9%
1.8%
2.1%
0.2%
3.2%
2.4%
2.4%*
2.4%*
2.4%**
4.8%***
6.4%***
None
8.7%
4.1%
3.4%
0.6%
0.6%
0.6%
0.9%
1.4
0.2%
6.1
10.3%***
15.1%***
0.8%
3.7%
71.4%***
0.2%
156
Table 3. FRAX risk factors and scores according to DXA scanning status.
Data are shown as medians (quartiles) or percentages.
Previous DXA
FRAX scores
Risk of major osteoporotic fracture
Risk of hip fracture
10-year risk of major osteoporotic fracture is ≥15%
10-year risk of major osteoporotic fracture is ≥20%
10-year risk of hip fracture is ≥10%
10-year risk of hip fracture is ≥15%
Number of FRAX risk factors
0
Falls and family history of osteoporosis included
1
2
3+
Falls and family history of osteoporosis included
*p<0.05, **p<0.01, ***p<0.001
Whole study
population
Yes
(n=126)
No
(n=4696)
(n=4696)
14.0 (11-18)***
4.6 (2.6-7.2)***
44.4%**
20.6%***
13.5%**
7.9%***
12.0 (9.1-16)
3.0 (1.9-5.3)
30.9%
10.3%
6.5%
2.2%
12.0 (9.1-16.0)
4.0 (1.9-5.4)
31.2%
10.6%
6.6%
2.4%
31.8%***
17.5%***
39.7%
21.4%***
7.9%***
23.8%***
52.5%
39.9%
36.1%
9.6%
1.8%
6.1%
51.9%
39.3%
36.2%
9.9%
2.0%
6.6%
157
Table 4. Uni- and multivariate logistic regression of the use of DXA in men.
R2:0.07
FRAX
Age (years)
BMI< 19 kg/m2
Family history of hip fracture
Fracture after the age of 50 years
Current smoker
Excessive use of alcohol
Current oral glucocorticoid treatment
RA
Secondary osteoporosis
Additional risk factors
Family history of osteoporosis
Falls within the last year
Socioeconomics
Income in quartiles
Q1
Q2
Q3
Q4
Married
Disability pension
Early retirement pension
1.05 (1.01-1.10)
2.04 (0.60-6.96)
1.24 (0.71-2.14)
2.72 (1.76-4.21)
1.59 (1.09-2.32)
1.08 (0.60-1.93)
8.90 (4.25-18.7)
2.60 (1.07-6.32)
3.98 (1.86-8.54)
Multivariate
OR (95%CI)
R2:0.09
R2: 0.09
1.05 (1.01-1.10)
1.86 (0.54-6.39)
1.09 (0.62-1.92)
2.34 (1.49-3.65)
1.59 (1.08-2.33)
1.04 (0.58-1.87)
8.83 (4.23-18.4)
2.58 (1.07-6.21)
3.86 (1.79-8.31)
1.05 (1.00-1.11)
1.75 (0.50-6.08)
1.09 (0.62-1.93)
2.26 (1.44-3.54)
1.51 (1.03-2.22)
1.05 (0.58-1.89)
8.67 (4.14-18.1)
2.45 (1.02-5.88)
3.61 (1.67-7.83)
1.64 (0.97-2.78)
1.99 (1.35-2.94)
1.66 (0.98-2.82)
1.93 (1.30-2.86)
R2:0.08
1.05 (1.01-1.10)
2.42 (1.55-3.77)
1.62 (1.11-2.37)
9.05 (4.35-18.8)
2.56 (1.07-6.15)
3.84 (1.79-8.23)
2.01 (1.36-2.96)
1
0.94 (0.54-1.63)
0.93 (0.55-1.56)
0.75 (0.42-1.34)
0.92 (0.59-1.43)
1.68 (0.87-3.23)
0.95 (0.55-1.64)
The columns represent regression models including 1) clinical risk factors used in FRAX, 2)
additional clinical risk factors, 3) additional socioeconomic factors, and 4) significant factors from
the preceding models.
158
Reference List
(1) van Staa TP, Dennison EM, Leufkens HG, Cooper C. Epidemiology of fractures in England and Wales. Bone 2001
Dec;29(6):517-22.
(2) Sambrook PN, Seeman E, Phillips SR, Ebeling PR. Preventing osteoporosis: outcomes of the Australian Fracture
Prevention Summit. Med J Aust 2002 Apr 15;176 Suppl:S1-16.
(3) Kiebzak GM, Beinart GA, Perser K, Ambrose CG, Siff SJ, Heggeness MH. Undertreatment of osteoporosis in men
with hip fracture. Arch Intern Med 2002 Oct 28;162(19):2217-22.
(4) Abrahamsen B, van ST, Ariely R, Olson M, Cooper C. Excess mortality following hip fracture: a systematic
epidemiological review. Osteoporos Int 2009 Oct;20(10):1633-50.
(5) MacLean C, Newberry S, Maglione M, McMahon M, Ranganath V, Suttorp M, et al. Systematic review:
comparative effectiveness of treatments to prevent fractures in men and women with low bone density or
osteoporosis. Ann Intern Med 2008 Feb 5;148(3):197-213.
(6) Schousboe JT, Taylor BC, Fink HA, Kane RL, Cummings SR, Orwoll ES, et al. Cost effectiveness of bone
densitometry followed by treatment of osteoporosis in older men. JAMA 2007 Aug 8;298(6):629-37.
(7) Vestergaard P, Rejnmark L, Mosekilde L. Osteoporosis is markedly underdiagnosed: a nationwide study from
Denmark. Osteoporos Int 2005 Feb;16(2):134-41.
(8) Abrahamsen B, Vestergaard P. Declining incidence of hip fractures and the extent of use of anti osteoporotic
therapy in Denmark 1997-2006. Osteoporos Int 2010 Mar;21(3):373-80.
(9) Kanis JA, Oden A, Johnell O, Johansson H, De Laet C, Brown J, et al. The use of clinical risk factors enhances the
performance of BMD in the prediction of hip and osteoporotic fractures in men and women. Osteoporos Int
2007 Aug;18(8):1033-46.
(10) Cummings SR, Cawthon PM, Ensrud KE, Cauley JA, Fink HA, Orwoll ES. BMD and risk of hip and nonvertebral
fractures in older men: a prospective study and comparison with older women. J Bone Miner Res 2006
Oct;21(10):1550-6.
(11) Ewald DP, Eisman JA, Ewald BD, Winzenberg TM, Seibel MJ, Ebeling PR, et al. Population rates of bone
densitometry use in Australia, 2001-2005, by sex and rural versus urban location. Med J Aust 2009 Feb
2;190(3):126-8.
(12) Jones G, Nguyen T, Sambrook PN, Kelly PJ, Gilbert C, Eisman JA. Symptomatic fracture incidence in elderly men
and women: the Dubbo Osteoporosis Epidemiology Study (DOES). Osteoporos Int 1994 Sep;4(5):277-82.
(13) Johnell O, Kanis JA. An estimate of the worldwide prevalence and disability associated with osteoporotic fractures.
Osteoporos Int 2006 Dec;17(12):1726-33.
(14) Geusens P, Dinant G. Integrating a gender dimension into osteoporosis and fracture risk research. Gend Med
2007;4 Suppl B:S147-S161.
(15) Morris CA, Cabral D, Cheng H, Katz JN, Finkelstein JS, Avorn J, et al. Patterns of bone mineral density testing:
current guidelines, testing rates, and interventions. J Gen Intern Med 2004 Jul;19(7):783-90.
(16) Papaioannou A, Kennedy CC, Ioannidis G, Gao Y, Sawka AM, Goltzman D, et al. The osteoporosis care gap in
men with fragility fractures: the Canadian Multicentre Osteoporosis Study. Osteoporos Int 2008
Apr;19(4):581-7.
(17) Kanis JA, Johnell O, Oden A, Johansson H, McCloskey E. FRAX and the assessment of fracture probability in
men and women from the UK. Osteoporos Int 2008 Apr;19(4):385-97.
159
(18) Neuner JM, Zhang X, Sparapani R, Laud PW, Nattinger AB. Racial and socioeconomic disparities in bone density
testing before and after hip fracture. J Gen Intern Med 2007 Sep;22(9):1239-45.
(19) Demeter S, Leslie WD, Lix L, MacWilliam L, Finlayson GS, Reed M. The effect of socioeconomic status on bone
density testing in a public health-care system. Osteoporos Int 2007 Feb;18(2):153-8.
(20) Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in
longitudinal studies: development and validation. J Chronic Dis 1987;40(5):373-83.
(21) Ryg J, Nissen N, Nielsen D, Brixen KT. [Patients' and population's knowledge of osteoporosis]. Ugeskr Laeger
2005 Jan 17;167(3):282-5.
(22) Gaines JM, Marx KA, Caudill J, Parrish S, Landsman J, Narrett M, et al. Older men's knowledge of osteoporosis
and the prevalence of risk factors. J Clin Densitom 2010 Apr;13(2):204-9.
(23) Doheny MO, Sedlak CA, Estok PJ, Zeller R. Osteoporosis knowledge, health beliefs, and DXA T-scores in men
and women 50 years of age and older. Orthop Nurs 2007 Jul;26(4):243-50.
(24) Otmar R, Henry MJ, Kotowicz MA, Nicholson GC, Korn S, Pasco JA. Patterns of treatment in Australian men
following fracture. Osteoporos Int 2010 Mar 13.
(25) Bliuc D, Eisman JA, Center JR. A randomized study of two different information-based interventions on the
management of osteoporosis in minimal and moderate trauma fractures. Osteoporos Int 2006;17 (9):1309-17.
(26) Rubin KH, Abrahamsen B, Hermann AP, Bech M, Gram J, Brixen K. Prevalence of risk factors for fractures and
use of DXA scanning in Danish women. A regional population-based study. Osteoporos Int 2010 Aug 4.
(27) Kanis JA, Johnell O. Requirements for DXA for the management of osteoporosis in Europe. Osteoporos Int 2005
Mar;16(3):229-38.
(28) Haapanen N, Miilunpalo S, Pasanen M, Oja P, Vuori I. Agreement between questionnaire data and medical records
of chronic diseases in middle-aged and elderly Finnish men and women. Am J Epidemiol 1997 Apr
15;145(8):762-9.
(29) Kehoe R, Wu SY, Leske MC, Chylack LT, Jr. Comparing self-reported and physician-reported medical history.
Am J Epidemiol 1994 Apr 15;139(8):813-8.
(30) Bush TL, Miller SR, Golden AL, Hale WE. Self-report and medical record report agreement of selected medical
conditions in the elderly. Am J Public Health 1989 Nov;79(11):1554-6.
(31) Paganini-Hill A, Chao A. Accuracy of recall of hip fracture, heart attack, and cancer: a comparison of postal survey
data and medical records. Am J Epidemiol 1993 Jul 15;138(2):101-6.
(32) Brennan SL, Henry MJ, Wluka AE, Nicholson GC, Kotowicz MA, Pasco JA. Socioeconomic status and bone
mineral density in a population-based sample of men. Bone 2010 Apr;46(4):993-9.
(33) Brennan SL, Pasco JA, Urquhart DM, Oldenburg B, Hanna F, Wluka AE. The association between socioeconomic
status and osteoporotic fracture in population-based adults: a systematic review. Osteoporos Int 2009
Sep;20(9):1487-97.
(34) Vestergaard P, Rejnmark L, Mosekilde L. Socioeconomic aspects of fractures within universal public healthcare: a
nationwide case-control study from Denmark. Scand J Public Health 2006;34(4):371-7.
`