番茄社区

ATLANTA -- Women who had type 2 diabetes for longer than 10 years were at increased risk for having major osteoporotic fractures as well as hip fractures, a Canadian study found.

Among women enrolled in the Manitoba Bone Mineral Density (BMD) cohort, those with diabetes duration exceeding 10 years had a hazard ratio for major osteoporotic fracture of 1.47 (95% CI 1.30-1.66), according to , of the University of Manitoba, and colleagues.

In addition, after adjustment for factors including comorbidity scores, falls requiring hospitalization, and use of osteoporosis therapies and insulin, the risk was attenuated but still significant (HR 1.34, 95% CI 1.17-1.54), Leslie reported at the American Society for Bone and Mineral Research annual meeting here.

Type 2 diabetes is associated with an increased risk for major osteoporotic and hip fractures than would be predicted with the World Health Organization's Fracture Risk Assessment (FRAX) risk assessment tool, he explained.

"We're aware from meta-analyses that diabetes paradoxically increases the risk for major osteoporotic fractures by 20% to 30%, and the risk of hip fractures by even more, by 70% to 80%. The FRAX tool, which is calibrated for the general population, does not perform particularly well in this population, underestimating the risk in patients with diabetes," Leslie said during his presentation.

"The mechanisms by which diabetes influences fracture risk are clearly complex and multifactorial, and include skeletal factors such as compromised bone strength and suppressed bone turnover, as well as nonskeletal factors such as the risk of falls and injury, obesity, and decreased visual acuity," he said.

"But all of these factors require time for the diabetic milieu to have adverse effects, and studies that have looked at the time effects of type 2 diabetes have been inconsistent, with some showing no association with increasing duration, others showing effects only with long duration, and others even suggesting a transient reduction in risk with new-onset diabetes.

Therefore, to evaluate the potential greater risks with longer duration of diabetes, the researchers identified women older than 40 who had at least 10 years of health coverage who underwent baseline hip dual emission x-ray absorptiometry (DXA) measurements from 1996 to 2013.

FRAX scores were calculated according to BMD at the femoral neck.

The BMD registry included 49,098 women without and 8,840 women with diabetes. In 2,776 women, the diagnosis of diabetes had been made before DXA testing.

In 31.4%, the duration of diabetes was longer than 10 years, in 20.1% the duration was 5 to 10 years, and in 23.7%, it was less than 5 years. In the remaining 24.8%, the diabetes diagnosis was made within 5 years after the DXA testing.

During a median follow-up of 7 years, there were 814 major osteoporotic fractures and 279 hip fractures.

There was no increased risk of major osteoporotic fracture for diabetes duration of less than 10 years.

In contrast, there were increased risks for hip fracture for all diabetes durations. For new-onset diabetes, the HR was already significant, at 1.32 (95% CI 1.03-1.69), rising to 2.10 (95% CI 1.71-2.59) after 10 years.

After adjustment for additional covariates, little attenuation was seen in the HRs for hip fracture.

The researchers also confirmed that FRAX scores significantly underestimated risks. When they calculated the calibration ratio of observed versus predicted events, they found that for those with diabetes duration longer than 10 years, the risk of fracture was 24% greater than predicted (1.24, 95% CI 1.08-1.39). For hip fractures, the ratio of observed to predicted was almost double (1.93, 95% CI 1.50-2.35).

"The clinical implications 0f this study are that FRAX underestimates risk, especially for hip fractures, in patients with long duration of diabetes, and raises the question of whether type 2 diabetes of long duration should be considered a red flag for greater attention, perhaps as a "prior fracture equivalent" much in the same way as long duration of diabetes can be considered a "coronary heart disease equivalent" when performing cardiac risk assessment," Leslie said.

Comment