Oral 63rd Endocrine Society of Australia Annual Scientific Meeting 2020

Fractures in T2DM independently predicted by insulin use and vascular complications (FIELD study) (#28)

Angela Sheu 1 2 3 , Rachel O'Connell 4 , Alicia Jenkins 4 , Jackie Center 1 2 3 , Christopher White 1 3 5 , Tony Keech 4
  1. Bone Division, The Garvan Institute, Sydney, NSW, Australia
  2. Endocrinology Department, St Vincent's Hospital, Sydney, NSW, Australia
  3. University of New South Wales Faculty of Medicine, University of New South Wales , Sydney
  4. NHMRC Clinical Trials Centre, University of Sydney, Sydney
  5. Department of Endocrinology and Metabolism, Prince of Wales Hospital, Sydney

Background: Type 2 diabetes mellitus (T2DM) is associated with increased risk of some fractures although the exact mechanisms are unclear. Bone fragility may be associated with T2DM severity (microvascular complications, longer duration, insulin use), although prospective studies evaluating their independent contributions are lacking.

Aims: To determine whether baseline micro- or macrovascular disease predict incident fractures in T2DM, and whether T2DM medications or clinical characteristics independently contribute to fracture risk.

Methods: Fenofibrate Intervention and Event Lowering in Diabetes (FIELD) was a randomised controlled trial of fenofibrate therapy in T2DM patients (aged 50-75), with fractures collected as adverse events. In this post-hoc analysis, cox proportional hazards models were used to determine fracture predictors from baseline data.

Results: Over 49,470 person-years, 137/6138 men and 143/3657 women suffered a fracture. Men with fracture (vs without) were older (63.7±7.5 vs 62.4±6.8 years, p=0.03), more likely to have macrovascular disease (32.9% vs 23.4%, p=0.01), use insulin (21.9% vs 13.6%, p=0.005) and had longer T2DM duration (8.1±6.9 vs 6.9±6.2 years, p=0.03). Women with fractures had more neuropathy (9.8% vs 4.3%, p=0.002) and greater insulin use (22.4% vs 13.3%, p=0.002). Age was similar in women with and without fracture (61.8±6.8 vs 62.9±7.5 years, p=0.06). Overall, mean HbA1c was 7.1%.

In men, significant univariate predictors for fracture included age, macrovascular disease, T2DM duration, insulin use and serum triglycerides, but only insulin use (HR 1.69 (1.12-2.54), p=0.01) and macrovascular disease (HR 1.47 (1.02-2.12), p=0.04) remained significant in multivariable modelling.

In women, significant univariate predictors included age, neuropathy, insulin use and serum creatinine, but only neuropathy (HR 2.16 (1.23-3.80), p=0.007) and insulin use (HR 1.65 (1.11-2.47, p=0.01) remained in multivariable modelling.

Conclusions: Insulin use and gender-specific vascular complications (macrovascular disease in men and neuropathy in women) predict fractures in T2DM. Studies evaluating insulin use and T2DM fracture risk are warranted.