From the Journals

FRAX with BMD may not be accurate for women with diabetes



The fracture risk assessment tool FRAX may underestimate fracture risk in women with diabetes when bone mineral density is included, according to data from 566 women aged 40-90 years.

Image of a fractured spine ©wildpixel/Thinkstock

In a study published in Bone Reports, Lelia L.F. de Abreu, MD, of Deakin University, Geelong, Australia, and colleagues investigated the accuracy of FRAX scores and the role of impaired fasting glucose (IFG) and bone mineral density (BMD) on fracture risk by comparing FRAX scores for 252 normoglycemic women, 247 women with IFG, and 67 women with diabetes.

When BMD was not included, women with diabetes had a higher median FRAX score for major osteoporotic fractures of the hip, clinical spine, forearm, and wrist than women without diabetes or women with IFG (7.1, 4.3, and 5.1, respectively). In the diabetes group, 11 major osteoporotic fractures were observed versus 5 predicted by FRAX. In the normoglycemic group, 28 fractures were observed versus 15 predicted, and in the IFG group 31 fractures were observed versus 16 predicted.

When BMD was included, major osteoporotic fractures and hip fractures also were underestimated in the diabetes group (11 observed vs. 4 observed; 6 observed vs. 1 predicted, respectively), but the difference in observed versus predicted fractures trended toward statistical significance but was not significant (P = .055; P = .52, respectively). FRAX with BMD increased the underestimation of major osteoporotic fractures in the normoglycemic and IFG groups (28 observed vs. 13 predicted; 31 observed vs. 13 predicted).

The study findings were limited by several factors including the inability to determine the impact of specific types of diabetes on fracture risk, lack of data on the duration of diabetes in study participants, the use of self-reports, and a relatively small and homogeneous sample size, the researchers noted.

However, the results support data from previous studies showing an increased fracture risk in diabetes patients regardless of BMD, and suggest that FRAX may be unreliable as a predictor of fractures in the diabetes population, they concluded.

The study was supported in part by the Victorian Health Promotion Foundation, National Health and Medical Research Council Australia, and the Geelong Region Medical Research Foundation. Two researchers were supported by university postgraduate rewards and one researcher was supported by a university postdoctoral research fellowship. The remaining coauthors reported no relevant financial conflicts.

SOURCE: de Abreu LLF et al. Bone Reports. 2019 Aug 13. doi: 10.1016/j.bonr.2019.100223.

Next Article: