OSTEOPOROSIS by Ingrid Nelson (9/16/2003)
In recent years,
drug companies have been aggressive about marketing osteoporosis as a disease
that their drugs can treat. For some
people, osteoporosis is a disease. However,
the thinning of bones is also a normal part of aging for both men and women, a
genetically programmed process that begins—along with a multitude of other
physiologic changes associated with growing older, rising blood pressure, for
example—after age 30 or so, and kicks into high gear at 50 years old.
In some people,
thin bones fracture; perhaps this is when osteoporosis becomes a disease. A corollary might be the unfortunate
individual with high blood pressure who suffers a stroke. Of course, not everybody with high blood
pressure has a stroke, just as not everybody with thin bones sustains a
fragility fracture. So, one way to
think about osteoporosis might be as a risk factor for fracture (just as high
blood pressure is a risk factor for stroke).
It’s probably
possible to predict such fractures and maybe even to prevent them. And, future fractures in a patient with the
disease of osteoporosis may also be preventable. So, as a disease, osteoporosis is treatable, and as a potential
disease, it may also be preventable.
Here’s the
obligatory rundown on the costs of osteoporitic fractures: it’s huge.
But consider also the pain, disfigurement, depression, and declining
productivity that the disease of osteoporosis inflicts on those it affects, and
I think you’ll agree that these numbers pale in comparison. A frequently quoted statistic that your
average white woman has a one in six chance of fracturing her hip during her
lifetime. She may have a one in two
chance of fracturing something.
A quick but
important note: most of the figures
I’ll cite were taken from studies of white, post-menopausal women living in
developed nations.
DEFINITION
Bone is
constantly being formed and reformed.
It’s a coupled process—first resorption and then formation. During the years of skeletal growth, the
amount of new bone formed exceeds the amount resorbed. There’s a big burst in growth during puberty
with a slower rate of growth after that until peak bone mass is reached—at
about the age of 30 or so. The
determinants of this mass are several.
Genetics are probably most significant, probably accounting for 50% of
the final bone mass reached. This may
be why white women are far more likely to become osteoporetic than African
American women. But, diet, exercise,
and co-morbid conditions also play a role (clearly an opportunity for early
intervention).
In any case, no
matter how terrific your bones are they begin to lose mass eventually. This happens because in each cycle, the rate
of resorption begins to overtake the rate of formation. In an adult, the coupled cycle can take a
year or so to complete, a useful thing to know when one considers how to
evaluate the success of therapy. In any
case, this gradual age-related decline accounts for about a one percent loss in
bone mass a year. This loss can be
accelerated by decreased activity, poor diet, increased bad habits (smoking,
drinking, etc.), and an age-related increase in co-morbid conditions with
increasing drug use. Women have a
additional period of bone-loss—the post-menopausal years when, due to a drop in
estrogen levels, they may lose 10-25% of their bone mass. This occurs in the 10 to 15 years following
menopause and is independent of age-related bone loss. Note that the loss of bone is not just
volumetric; bony architecture is also affected. The trabeculae that support the mineralized bone are weakened,
and fragility fractures become more likely.
Now, while all
women lose bone after menopause, not all (perhaps 20% of white women) become
osteoporitic. It’s unclear why they do,
but, again, the reasons are probably several—lower peak bone mass, lower
post-menopausal estrogen levels (or higher levels of sex hormone binding globulins),
or impaired bone formation processes.
The chief risk
factors for getting osteoporosis (as summarized in a review article in the NEJM
in 1998) are:
Heredity
First-degree relative with a low trauma
fracture
Lifestyle
Cigarette
smoking
Alcohol
abuse
Physical
inactivity
Thin
habitus
Diet low in
calcium
Little
exposure to sunlight
Menstrual status
Early
menopause (before 45 yrs. old)
Previous
periods of amenorrhea (at least one year)
Drugs
Chronic
steroid use
Anti-epileptic
drugs
Excessive substitution
therapy (i.e., thyroxin, corticosteroids)
Anti-coagulant
drugs
Endocrine diseases
Hyperparathyroidism
Thyrotoxicosis
Cushing’s
syndrome
Addison’s
disease
Hematological
diseases
Rheumatologic
diseases
Rheumatoid
arthritis
Ankylosing
spondylitis
GI diseases
Malabsorption
*Some drugs reduce calcium retention—common ones are INH,
heparin, tetracyclines, and lasix.
In any case, by
age 80 women have lost on average 40% of their peak bone mass and men have lost
about 25%. Moreover, the bone that
remains has distorted architecture, a synergistic effect to absolute loss of
mass that further contributes to structural weakness.
Now, the loss of
bone mass isn’t a bad thing per se.
It’s only bad when the bone breaks.
And, that can be terrible. You
are all probably aware of the morbidity and mortality associated with hip
fractures, with the chronic pain that comes with vertebral fractures, the
inconvenience associated with wrist fractures—the three types most commonly
associated with osteoporosis.
How often does
osteoporosis lead to fracture?
Impossible to say, but here is a chart of the lifetime risk of various
fractures at age 50 in this country:
Vertebral
Hip Forearm Clinical Radiographic
White Women
17% 16% 16% 35%
White Men
6% 3% 5% ?
Black Women
6% ? ? ?
Black Men
3% ? ? ?
What risk
factors make one person more likely to sustain a non-traumatic fracture than
another? A 1995 survey of ambulatory
women, average age 72, and almost all white, identified 16 independent ones;
those that are starred were most significant:
Age
*History of
maternal hip fracture
No increase
in weight since age 25
Taller
Self-rated
health
*Previous
hyperthyroidism
Current use
of benzodiazepines
*Current
use of anti-convulsant drugs
High
current caffeine intake
Not walking
for exercise
On feet
less than four hours a day
*Inability
to rise from a chair
Poor
distant depth perception
Poor
contrast sensitivity
*Resting
pulse greater than 80
Any
fracture since age 50
(Other studies have included smoking)
Note some
overlap between the two lists, and note also that all of the above fit into
four basic categories:
Heredity
Lifestyle
Medical conditions
Falls (or fate)
The two most
readily modifiable are, obviously, lifestyle and falls. You’ll hear more about falls in another
talk; suffice it to say that most hip fractures are a result of falls. And, with a bit of consideration, you can
see that low bone mass density is a final common pathway in all three groups.
Until the late
1980’s it was difficult to measure bone density accurately. However, most hospitals now have dual energy
x-ray absorptiometry machines that, with a dose of radiation just a tenth of
that of a regular chest x-ray, can measure bone density accurately to within a
few percentage points. With the
development of these machines it became possible to augment the way women are evaluated
for risk, and to see which treatments work.
Instead of using just fracture as an endpoint, which involved following
very large groups of people for many years, it was possible to design
prospective, randomized studies to see how one treatment or another affect bone
density and, thus, presumably, risk of fracture.
Note that BMD
readings give a picture of how the bones are now. There are certain biochemical markers that may be more
predictive; that is, give a picture of how the bones may be in years to
come. None have been rigorously tested,
but one in common use is N-telopeptides, a fairly specific marker for bone
breakdown that appears in the urine.
Elevated levels suggest that bone is being resorbed and, when patients
are of the age where rates of resorption exceed rates of formation, such levels
may suggest that the patient is headed for osteoporosis, or at least is at
higher risk of getting there. So some
centers use these markers prospectively in patients at risk for the
disease.
Now the use of BMD in studies isn’t ideal for
our purposes. In making clinical
decisions it’s best to rely on studies that look at clinical outcomes rather
than surrogates. However, BMD is highly predictive of the risk of fracture; in
fact the predictive power is stronger than that of hypertension for stoke, or
of hyperlipidemia for coronary artery disease.
BMD readings have given rise to a definition of osteoporosis, based on
the so-called T-score, or a comparison of any set of readings to those for a mean
for young women (or man) which is arrived at by manufacturers by pooling
readings from a large base of racially diverse people (divided by gender). DEXA readings are also graded by a Z-score,
which compares your patient’s reading to age matched controls.
The WHO has
proposed three categories:
1.
Osteopenia, with BMDs between 1 and 2.5 SD below the mean.
2.
Osteoporosis, with BMD values more than 2.5 below the mean.
3.
Severe osteoporosis, with BMD values at least 2.5 SD below the
mean and a history of nonviolent fracture.
Any person who has sustained a fragility fracture has
osteoporosis.
Now keep a few
things in mind. Low BMD is a risk
factor for a fracture, not a fracture itself, and the risk of having a disease
(that is, osteoporitic fractures) is not the same as having the disease. Low BMD is specific, but not especially
sensitive, in predicting fracture.
Osteo-arthritis may give falsely high BMDs; one way around this is to
measure the hip not at the joint but at the femoral neck. Finally, older women are at higher risk for
fracture than younger women with a comparable BMD.
Measurement at
any site is predictive, but measurement of the trochanter is most accurate in
predicting a hip fracture, and measurement at the vertebrae is best to assess
treatment efficacy.
What do these
numbers mean in terms of bone mass?
Well, a decline of one standard deviation below the mean at the spine
represents a loss of bone mass of about 10%.
How about in terms of fracture? Well,
the risk of fracture increases about 2 times for each decrease of one SD.
WHO GETS STUDIES?
NEW FLASH: As of the
new issue of the US guide to preventative services, screening for osteoporosis
is recommended for all women over 65 yo, and for women between 60 and 65 yo
with risk factors.
Those with risk
factors. Those considering
pharmacological treatment. Those with
clinical signs of osteoporosis—one of the most common is excessive loss of
height (greater than two inches); also those with non-traumatic fractures of
the wrist, hip, or vertebrae.
Note that, in
addition to a DEXA scan, your work-up should also include thyroid tests, a PTH,
and urine calcium to diagnose hypercalciuria.
You might also consider protein electrophoresis and cortisol levels, if
indicated.
DEXA is a good
way to follow treatment, too. But,
because of the long bone remodeling cycle, don’t repeat the test more
frequently than each 18 months or so.
WHO GETS PHARMOCOLOGIC TREATMENT?
Well, everyone
in whom it is not contra-indicated should take calcium (1200 to 1500 mg per
day) and Vitamin D (400 to 800 IU per day).
Everyone should be counseled about risk factors. And everyone, present company not excluded,
should get more exercise and try to eliminate other risk factors like smoking
and drinking too much.
The National
Osteoporosis Foundation recommends pharmacological treatment of all
post-menopausal women with T scores below –2 and those with T scores below –1.5
and risk factors for osteoporosis. But
remember that at the same level of BMD the fracture risk is much greater in
older than in younger subjects. Also,
the greatest benefit in fracture reduction occurs in patients with preexisting
vertebral fractures or T scores below –2.5.
So, what’s
available and how well does it work?
Note that all approved agents prevent bone loss; none significantly
increase bone mass. That is, they act
by decreasing resorption. Since
formation is coupled to resorption, this is decreased also. But, by decreasing the number of cycles in
patients whose formation is out distanced by resorption overall bone density
tends to increase slightly—by 10% or so—then stabilize after three years or
so. The most effective of these agents
reduce fracture rates by about 50%.
- Calcium
and vitamin D: Supplementation may
increase bone density and decrease fracture rates, dependant, in part, on
the nutritional status of the patient.
The nutritional status of most American patients, at least in terms
of calcium, is pretty poor. The
average adult here takes in less than 600 mg. of calcium a day. There is an age-related
decrease in vitamin D receptors in the intestine and the hydroxylation of
vitamin D to its active form also decreases with age. So, there is a consequent decrease in
the amount of calcium absorbed.
Now, this gives a double whammy in terms of osteoporosis—lower
calcium levels stimulate PTH release and this, in turn, increases bone
turnover. (Note that pulsed PTH
actually increases bone mass, but we’re talking here about the steady
state levels.) This turnover
actually reduces bone mass, probably because there is also an age-related
decrease in the function of osteoblasts.
Several studies have shown that appropriate levels of calcium
intake (1500 mg. per day for post menopausal women and older men) increase
bone density and may reduce fracture rates. This should be supplemented, especially in winter, with
vitamin D (400 to 800 IU per day).
No side effects have been noted—note that with increasing levels of
calcium intake, the percent absorbed actually decreases, so it’s hard to
reach pathologic levels. Also note
that calcium and vitamin D should be taken by women and men over 30
(depending on their diet), and as an adjunct to other therapies. The only exception might be in early
post-menopausal women who have high calcium levels because of increased
bone breakdown. A final word: if you have patients who are younger
than 30, you might want to make sure they take in adequate calcium so that
they achieve a maximum bone mass. - Estrogen: Supplementation reduces all types of
fractures in post-menopausal women.
Estrogen affects bone density in several ways—by increasing the
activity of osteoblasts, decreasing the activity of osteoclasts, and
increasing calcium absorption.
It’s especially useful in the post-menopausal years, and especially
effective in women over 65. So,
timing of therapy might be important as, when therapy is discontinued, a
state akin to post-menopausal bone loss ensues. Low dose estrogen (0.3 mg per day) has been used to prevent
and treat osteoporosis, but seems to be less effective than the standard
dose of 0.625 mg. Some side
effects are fewer; it is unclear if the risk of cancer is decreased. (Note that there are no large RCTs evaluating
HRT for osteoporosis. The
information comes from observational studies, like the Framingham study. - Selective
estrogen receptor modulators (Raloxifene): Supplementation reduces fracture rates, but not as much as
estrogen. Choose this based on
other effects: estrogen reduces
hot flashes while Raloxifine doesn’t; Raloxifene reduces estrogen receptor
positive breast cancers while estrogen increases the risk of getting one;
Raloxifene does not increase the risk of endometrial cancer; both increase
the rate of thrombosis. Note that
serms have not been shown to reduce the rates of non-vertebral fractures. - Bisphosphanates: Increase BMD and decrease fracture
rates in patients with osteoporosis.
Medications like Fosomax bind to the surface of bone and inhibit
the activity of osteoclasts.
Osteoblasts are not inhibited.
Fosomax or risidronate (Actinol) can be used in conjunction with
estrogen if either therapy alone fails.
Fosomax is the best studied of the agents (although Actinol is less
irritating to the GI tract). It
can be given daily or once a week, and must be taken carefully to avoid
esophageal irritation. After BMD
readings level off (usually after 4 or 5 years of therapy), it should be
given on a more intermittent basis (every other year or every other
day). These medications are
relatively new, and long-term side effects are yet to be determined,
although they have been used for many years for other purpose. Note that their protective effect on
bone is probably fairly long lasting, unlike hormones, whose protective
effect disappears as soon as they are stopped. - Calcitonin: Probably reduces fracture rates, but
provides analgesia. The newish,
intra-nasal formulation is easy to take.
This is a good choice for women who can take neither estrogen nor
Fosomax, and particularly useful for those with back pain due to vertebral
fractures. I know of no studies
that have looked at this in combination with other agents.
FUTURE DIRECTIONS
A promising
therapy seems to be pulsed PTH injections.
Unlike the above treatments, this may significantly increase bone
mass. As yet, it is a study drug.
A FINAL NOTE
Most hip
fractures occur after the age of 65.
The treatments outlined above (with the exception of hormone replacement
given at menopause to prevent post-menopausal bone loss) work no matter when
they are started. So, in people at
average risk, you might consider recommending beginning treatment when their
risk of fracture is greatest—that is, when they are older. It’s a way to improve your NNT. This said, everyone should take supplemental
calcium and vitamin D.
A FINAL FINAL NOTE
An eight-ounce
glass of milk contains about 300 mg. of calcium. An eight-ounce container of yogurt contains about 400 mg. of
calcium. One ounce of cheese has about
200 mg. of calcium. And, in the average
American diet, non-dairy sources of calcium account for about 250 mg. a day. We’re talking spinach.
