When I think of minimal trauma fractures — fractures occurring
after a fall from a standing height or lesser trauma — I think of
osteoporosis, both primary and secondary. When it comes to fractures in
patients with type 2 diabetes I don’t know what to think, yet I am
encountering this in the clinic with increasing regularity. I spent several
hours on the web trying to find out more but aside from confirming the
epidemiology, the prediction, etiology and management remains unclear.
What is clear is that patients with type 2 diabetes have a higher bone
mineral density than age-, sex- and ethnicity-matched patients without
diabetes, even when a higher BMI is taken into consideration; those who sustain
a minimal trauma fracture do so with a higher average BMD at all measurement
sites than patients without diabetes who sustain similar fractures; fractures
occur at the common sites of osteoporosis-related fractures — spine, hip,
wrist and ribs; fractures are more common at unusual sites such as the foot and
ankle; and therapy with thiazolidinediones (TZDs) is an apparent additional
risk factor for fracture but we cannot, to my knowledge, risk stratify our
patients on TZDs.
Some other associations are obvious. If your patient has episodes of
hypoglycemia it should be no surprise that any sudden and uncontrolled fall may
result in fracture; similarly with a misstep in patients with peripheral
neuropathy or patients who have gait disturbance after a stroke. Preventive
measures in these circumstances are straightforward, minimize the likelihood of
hypoglycemia and provide walking aids for those with neuropathy or post-stroke.
As for the rest, which is the majority of patients:
There appears to be an association between pentosidine levels and risk
of fracture. Pentosidine is a biomarker for advanced glycation end products
(AGEs). Several reliable assays are available for this and the higher the level
of pentosidine the greater the likelihood of fracture. Basically this means the
worse the glycemic control in type 2 diabetes the greater the risk for fracture
— as with all other complications of type 2 diabetes.
Low levels of adiponectin and a low adiponectin to high-molecular-weight
adiponectin ratio have also been linked epidemiologically to an increased risk
for fracture. Adiponectin is a member of a large family of protein hormones and
cytokines secreted by fat cells known collectively as adipokines. In the
circulation there is both adiponectin and high-molecular-weight adiponectin,
and assays are available for both. Adiponectin is an important regulator of
insulin sensitivity and, seemingly paradoxically, production of adiponectin is
reduced in obesity. Type 2 diabetes is characterized by low levels of
adiponectin and a lower ratio of adiponectin to high-molecular-weight
adiponectin. Once again, the worse the control of type 2 diabetes, the more
likely are the complications including fractures.
From my reading it is unclear that either pentosidine or adiponectin are
better markers of long-term control of diabetes than HbA1c but the clinical
implementation of these assays is still limited and it would be no great
surprise if over time they replace HbA1c or serve as adjuncts to HbA1c to help
us and our patients maintain better control of their disease.
Important as all of the above might be in theory, none of it is really
yet helpful in making a decision about which of our patients might be at
greatest risk for fragility fracture and should be considered for prophylaxis
to prevent fracture.
As already noted, BMD does not help too often. What about biochemical
markers of bone turnover? The jury is still out but I am increasingly measuring
them in patients who I am seeing in clinic for their diabetes, not their bones.
My experience is still small but most have values within the reference
interval, and the few with elevated levels have an identifiable non-diabetes
related cause such as vitamin D deficiency. When the bone turnover levels are
OK I cannot justify recommending anything but supplemental calcium and vitamin
D, optimize glycemic control and minimize fall risk.
Then there are a few in whom the markers of resorption and/or formation
are below the lower limit of the reference interval or close to it. They really
trouble me because I am convinced that oversuppression of bone remodeling is
not a good thing but there are no therapeutic options. Teriparatide is the only
FDA-approved formation stimulation option. Good luck tying to get insurance
coverage for this without first trying one of the other FDA-approved therapies
for osteoporosis, which in my mind are contraindicated. In fairness, there was
no statistical relationship between baseline markers and anti-fracture
effectiveness in any of the clinical trials of the FDA-approved anti-resorptive
therapies, but I don’t know of any specific data in patients with
suppressed remodeling at baseline.
Where does this leave us? Just another reason to do our best to work
with our patients about how to optimize control of their diabetes — and I
didn’t need the hours of literature-searching and two pages of typing to
arrive at this conclusion. But, I will close with a short list of suggested
Khazai NB. Curr Opin Endocrinol Diabetes Obes. 2009;16:435-445.
Yamamoto M. Diabetes Care. 2009;32:2263-2268.
Lecka-Czernik B. Curr Opin Investig Drugs. 2009;10:1085-1090.
Petit MA. J Bone Miner Res. 2009; Jul 13.
Schwartz AV. J Clin Endocrinol Metab. 2009;94:2380-2386.
Kanazawa I. J Endocrinol. 2009;160:265-273.
Melton LJ III. J Clin Endocrinol Metab. 2008;93:4804-4809.
Melton LJ III. J Bone Miner Res. 2008;23:1334-42.