Category: Central DXA: (DXA, TBS)

20 - Type 1 Diabetes and Bone Microarchitecture Assessment with Trabecular Bone Score (TBS): A Descriptive Study

Introduction: Type 1 Diabetes Mellitus (T1DM) is associated with increased risk of fragility fracture (FF), which is in excess of what is expected based on bone mineral density (BMD) alone. Impairment in bone quality may contribute to this excess risk. Trabecular bone score (TBS), a tool to assess bone quality at the lumbar spine (LS), has been shown to predict fracture risk independent of BMD. There are limited studies exploring TBS in T1DM.

Objective: To examine TBS in a cohort of adults with T1DM in relation to BMD, clinical parameters and fracture history.

T1DM patients were prospectively recruited from the Endocrine clinic at Women’s College Hospital from April to October 2015 for BMD testing with TBS. Clinical parameters and fracture history were obtained from survey and chart review. Patients on bone-sparing drugs or steroids were excluded. Descriptive statistics, correlation coefficients, multivariate linear and logistic regression models were used to determine associations between clinical variables, BMD, TBS and FF.

43 patients with T1DM participated in the study. Average age was 34.9 years, 70% female, 88% Caucasian. 44% were diagnosed with T1DM prior to puberty and average duration of T1DM was 16.7 years. Diabetes complications were uncommon; 1 had macrovascular disease, 26% had at least 1 microvascular complication. Average HbA1C was 8.2%. Almost half had history of fracture; of these, 33% were FF’s. Average BMD was 1.17 g/cm2 at LS, 0.94 g/cm2 at femoral neck (FN), 0.97g/cm2 at total hip (TH). Average TBS was 1.47 (SD 0.10). TBS correlated moderately with BMD; LS 0.41 (p= 0.006), FN 0.492 (p=0.001), TH 0.472 (p= 0.002). TBS did not correlate with age, gender, HbA1C, pre-pubertal DM onset, or duration of T1DM. BMI was the only variable that appeared to influence TBS. There was an inverse association between TBS and any microvascular complication, however this lacked statistical significance. The odds of FF increased with older age (OR 1.077, p=0.014) and decreased with higher BMD at the hip sites (OR 0.001, p= 0.029 at FN; OR 0.002, p=0.036 at TH). TBS did not predict FF.

TBS correlated with BMD at all sites and appears to be influenced by BMI, however there was no significant relationship between TBS, diabetes parameters and FF in this study cohort of young adults with T1DM. A larger study in older patients with longer duration of T1DM is required to better assess the utility of TBS in this patient population.

Anita Colquhoun

Head Technologist
Women's College Hospital
Toronto, Ontario, Canada

Julie Gilmour

University of Toronto, St. Michael's Hospital
Toronto, Ontario, Canada

Wei Wu

Research Associate
Women's College Research Institute
Toronto, Ontario, Canada

Sandra A. Kim

Endocrinologist, Assistant Professor
University of Toronto, Women's College Hospital
Toronto, Ontario, Canada