Association Between Glycemic Control and Risk of Fracture in Diabetic Patients

A Nested Case-Control Study

Janina Vavanikunnel; Sarah Charlier; Claudia Becker; Cornelia Schneider; Susan S. Jick; Christoph R. Meier; Christian Meier


J Clin Endocrinol Metab. 2019;104(5):1645-1654. 

In This Article

Abstract and Introduction


Context: Diabetes mellitus (DM) has been associated with an increased risk of fractures. However, the effect of glycemic control on the risk of fracture is not well understood.

Objective: To evaluate the association between glycemic control and the risk of low-trauma fractures in patients with type 1 DM (T1DM) and type 2 DM (T2DM).

Design: Nested case-control analysis.

Setting: UK-based Clinical Practice Research Datalink.

Patients or Other Participants: The study population was patients whose T1DM or T2DM had been newly diagnosed between 1995 and 2015. The cases were patients with a low-trauma fracture after DM onset. We matched four controls to each case by age, sex, general practice, fracture date, and DM type and duration.

Statistical Analysis: Conditional logistic regression analyses were performed, adjusted for covariates, including body mass index, smoking, DM complications and medications.

Results: The study population included 3329 patients with T1DM and 44,275 patients with T2DM. The median duration between DM onset and fracture date was 4.5 years for both T1DM and T2DM. The risk of fracture was increased in the patients with T1DM with a mean hemoglobin A1c >8.0% (adjusted OR, 1.39; 95% CI, 1.06 to 1.83) compared with those patients with T1DM and a mean hemoglobin A1c ≤7.0%. No such effect was found in the patients with T2DM. Independently of glycemic control, the risk of fracture was elevated in patients with T2DM and the current use of rosiglitazone and pioglitazone.

Conclusions: The effect of glycemic control on the risk of low-trauma fracture differs between patients with T1DM and T2DM. Poor glycemic control increased the risk of fractures in patients with T1DM but not in those with T2DM.


Diabetes mellitus (DM) has been associated with an increased risk of fragility fractures. In particular, the risk of hip fractures is increased approximately sixfold in subjects with type 1 DM (T1DM) and two- to threefold in patients with type 2 DM (T2DM).[1]

The pathophysiological mechanisms that contribute to skeletal fragility differ across the two DM types.[2] The differences in patient age at disease onset, insulin availability (insulin deficiency vs insulin resistance), and the influence of antidiabetic drugs lead to altered skeletal fragility. In patients with a diagnosis of T1DM during adolescence and early adulthood, the deficiencies in insulin and IGF-1 seem to impair osteoblast function, leading to lower bone mass, smaller bone size, and alterations in bone microstructure.[2–5]

In contrast, patients with T2DM, who usually experience obesity-related insulin resistance and hyperinsulinemia, will present with normal to increased bone mass and preserved or even increased trabecular bone volume but with increased cortical porosity. This pattern has been found especially in patients with fractures and microvascular complications.[3,6]

In patients with T1DM and T2DM with advanced disease, glucotoxicity, chronic inflammation, and microvascular changes have been thought to be critical factors in accelerating bone aging and the progression of diabetic bone disease.[7,8] In addition, nonskeletal factors,[9,10] such as chronic diabetic complications, comorbidities, and drug effects,[11,12] could increase the risk of falls[6,13] and, thus, the overall risk of fracture.

It remains unclear to what extent glycemic control has an impact on the risk of fracture. Some studies have reported an association between poor glycemic control and an increased risk of fractures[14–17] or falls,[18] but others have not.[1,19–21] In contrast, good glycemic control was also associated with an increased risk of fracture[13,19] or falls[22] in some studies. Although several studies have examined the effect of glycemic control on the risk of fractures by analyzing hemoglobin A1c (HbA1c) levels (a measure of the average glycemia during an ~12-week period), the results have remained inconsistent owing to methodological heterogeneities. Only a few of these studies included a substantial number of patients[14–16] and analyzed the effect of glycemic control, not only using a single HbA1c measurement,[14,22,23] but using the mean HbA1c levels during a longer follow-up period.[14] The latter might more accurately reflect the degree of glycemic control. Additionally, only Forsén et al.[1] analyzed the association between glycemic control and the risk of fracture separately for patients with T1DM and those with T2DM.

Therefore, we conducted a study to evaluate the association between the degree of glycemic control and the risk of nonvertebral low-trauma fractures in patients with newly diagnosed T1DM and T2DM.