Type 2 diabetes and fractures: more information is needed

Abstract

Type 2 diabetes mellitus (T2D) is an exceedingly common chronic metabolic disorder that has an enormous impact on public health. Currently, T2D affects over 366 million adults worldwide and is projected to reach 552 million by 2030 [1]. Until recently, the list of target organs affected by T2D did not include the skeleton. Yet there is now substantial evidence that older adults with T2D have a higher risk of sustaining a hip (RR = 1.7) or any clinical (RR = 1.2) fracture [2], morbidities that have important ramifications given the epidemic rates of this disease. Surprisingly, despite the increased fracture risk, bone mineral density (BMD) by dual energy X-ray absorptiometry (DXA) is generally higher in those with T2D [3]. In an analysis of three large prospective cohorts, BMD reliably predicted fracture risk in T2D, yet diabetics fractured at a T score that was higher by 0.6 than that of controls [4]. Similarly, two recent reports have shown that for a given fracture risk assessment (FRAX) probability, the risk of fracture among individuals with diabetes is higher than the risk in non-diabetics [4, 5]. Thus, although BMD and FRAX are able to predict fractures in T2D, both underestimate the actual fracture risk. This minimization may be partly explained by more frequent trauma, as diabetes is associated with an increased frequency of falls. However, even in studies of diabetes and fracture that control for fall frequency, diabetes remains independently associated with increased fracture risk [6]. Thus, the paradox of increased fracture risk with high bone density remains unresolved. Because BMD and FRAX are central to fracture prediction, a consequence of this paradox is a lack of suitable methods to predict fracture risk in older adults with T2D. In this issue of the Journal, Di Somma et al. [7] seek to address this issue by measuring the spinal deformity index (SDI), an index of fracture severity based on lumbar spine X-rays, in a group of 56 male and female T2D patients and 56 male and female controls. Consistent with prior reports, they found that morphometric vertebral fractures were more common in T2D than in controls (46 vs 17 %) and that more of the T2D patients fractured at T scores above -2.5 than did the controls (69 vs 10 %). The SDI measurement, a combination of the number of vertebral fractures and of the semiquantitative vertebral deformity scoring system of Genant et al. [8], was higher in T2D than in controls. The study concludes that the SDI measurement might be more useful than BMD for assessment of skeletal health in T2D. The authors should be commended for carefully examining the spine fracture burden and its characteristics in T2D patients. In particular, their findings are strengthened by the BMI matching between T2D and controls, which eliminates the potentially confounding skeletal effects of obesity. Moreover, the examination of calciotropic indices suggests that, regardless of BMI, vitamin D levels are lower and PTH levels are higher in diabetics. Nevertheless, the study is hampered by certain notable limitations. Importantly, the BMD scores were actually lower in the diabetics at both the lumbar spine and femoral neck, arguing against the point that the SDI measurement is providing fracture risk information that is lacking from BMD alone. Moreover, the study has not addressed the M. R. Rubin (&) Columbia University College of Physicians & Surgeons, Columbia University, PH8W-864, 630 W. 168th St., New York, NY 10032, USA e-mail: mrr6@columbia.edu