Osteoporosis is a major cause of morbidity and mortality among the
elderly and is all too often unrecognized in the pre-fracture state. At first
blush, this is difficult to understand because DXA predicts fracture at least
as well as blood pressure predicts a cerebrovascular accident and probably
better than total cholesterol predicts acute myocardial infarction. Whats
more, FDA-approved therapies for management of osteoporosis are held to a very
high standard since prevention of fractures, not just improvement in bone
mineral density, is required before any such therapy can be approved. To my
knowledge such a high standard (prevention of an adverse outcome) is not
required for approval of cholesterol or BP-lowering therapies.
The current issue of the Journal of Bone and Mineral
Research has an article1 and an accompanying editorial 2
that continue the uncertainty about how best to determine which patient
is most in need of intervention.
By now, you should all be familiar with the FRAX tool that was developed
by the WHO as a means of quantifying the 10-year probability of a hip fracture
and other major osteoporosis-related fractures. Cost-effective analysis
suggests cut points for fracture risk above which intervention to minimize
fracture risk becomes cost-effective. In the United States, the cut point for
hip fracture risk is a 10-year probability of 3%, and for other major
fractures, the cut point is 20%.
Using data from one of the earliest randomized controlled clinical
trials of alendronate (the FIT trial) a number of different analyses were
performed to see if FRAX was better than other models at predicting the
likelihood of vertebral fractures. The conclusion was that a combination
of baseline radiographic vertebral fracture, femoral neck BMD and age is the
strongest predictor of future vertebral fracture. The accompanying
editorial suggested that at first glance these studies seem to strike a
blow against FRAX, but on further review they almost
overturned the original ruling on the field. In the end, they concluded,
Will it (the new model) outperform FRAX in this narrow domain? Likely.
Will the public be better off with site-specific fracture assessment? We are
Shakespeare wrote about this Much ado about nothing.
Without detracting from the quality of the science and discussion in the
article and the editorial they are both first rate, and it is worth
getting hold of a copy from your library I just dont get the
message at all. In my practice, and I hope in yours, if a patient has a minimal
trauma vertebral fracture, intervention to prevent the next vertebral or other
osteoporosis-related fracture is indicated without regard to any formulae.
The question that should be addressed is how important is it to look for
silent vertebral fractures in women aged 65 or older and men aged 70 or older
who are scheduled for their first DXA? This can be done at the time of the DXA
if the instrument has the appropriate vertebral fracture assessment technology.
The method is quick, reliable and to my knowledge exposes the patient to less
ionizing radiation than sending them to the radiology department for a full set
of thoracic and lumbar spine X-rays.
There is also an article that addresses the timing of repeat BMD
measurements3, a vexed issue among patients who really want to know
ASAP that therapy is working. The authors had a large cohort with serial BMD
and fracture data and used a 10% risk model (either sustaining a fracture or
moving from the low bone mass [osteopenia] to the osteoporosis BMD level). Not
surprisingly if you are an 80-year-old woman with a femoral neck T-score of
2.2 at baseline, you would be likely to have a T-score of 2.5 or
lower if the DXA was repeated 1.5 years later. Equally unsurprisingly, a
60-year-old man with a baseline T-score of 0 could wait 10 or more years before
a repeat DXA would put him in the osteoporosis category.
The important message I gleaned from the article is that we really need
to individualize patient care, and these colleagues point us in the direction
as to how it can be done in the field of osteoporosis. That should be the
standard for each and every patient for whom we are providing care, no matter
what their disease or condition. Patients with marked dyslipidemia, severe
hypertension and uncontrolled diabetes, etc, need more frequent monitoring
until we are comfortable with the direction they are taking. It is nice to have
a documented model as this article provides. It would be even nicer if we had
that freedom to use our discretion and judgment with every patient.
For more information:
1. Donaldson MG. J Bone Miner Res. 2009;24:1793-1799.
2. Leslie WD. J Bone Miner Res.
2009;24:1789-1792. PMID: 19839767.
3. Frost SA. J Bone Miner
Res. 2009;24:1800-1807. PMID: 19419321.