High bone mineral density and fracture risk in type 2 diabetes as skeletal complications of inadequate glucose control: the Rotterdam Study

Abstract

Objective: Individuals with type 2 diabetes have increased fracture risk despite higher bone mineral density (BMD). Our aim was to examine the influence of glucose control on skeletal complications.

Research design and methods: Data of 4,135 participants of the Rotterdam Study, a prospective population-based cohort, were available (mean follow-up 12.2 years). At baseline, 420 participants with type 2 diabetes were classified by glucose control (according to HbA1c calculated from fructosamine), resulting in three comparison groups: adequately controlled diabetes (ACD; n = 203; HbA1c <7.5%), inadequately controlled diabetes (ICD; n = 217; HbA1c ≥ 7.5%), and no diabetes (n = 3,715). Models adjusted for sex, age, height, and weight (and femoral neck BMD) were used to test for differences in bone parameters and fracture risk (hazard ratio [HR] [95% CI]).

Results: The ICD group had 1.1-5.6% higher BMD, 4.6-5.6% thicker cortices, and -1.2 to -1.8% narrower femoral necks than ACD and ND, respectively. Participants with ICD had 47-62% higher fracture risk than individuals without diabetes (HR 1.47 [1.12-1.92]) and ACD (1.62 [1.09-2.40]), whereas those with ACD had a risk similar to those without diabetes (0.91 [0.67-1.23]).

Conclusions: Poor glycemic control in type 2 diabetes is associated with fracture risk, high BMD, and thicker femoral cortices in narrower bones. We postulate that fragility in apparently "strong" bones in ICD can result from microcrack accumulation and/or cortical porosity, reflecting impaired bone repair.

Figure 1

Adjusted means of narrow neck width (A) and cortical thickness (B) in relation to glucose control by age tertiles: youngest, 55.0–63.6 years of age; middle, 63.6–71.4 years of age; oldest, >71.4 years of age. Kaplan-Meier curve per comparison group showing the adjusted cumulative hazards for fracture using follow-up time as timescale (C). Cox proportional hazard model: ICD vs. no diabetes HR 1.47 (95% CI 1.12–1.92), P = 0.005; ACD vs. no diabetes HR 0.91 (0.67–1.23), P = 0.54. Cumulative HR adjusted for femoral neck BMD, age, sex, height, and weight. Light gray, ND; dark gray or dashed, ACD; black, ICD.

Figure 1

Adjusted means of narrow neck width (A) and cortical thickness (B) in relation to glucose control by age tertiles: youngest, 55.0–63.6 years of age; middle, 63.6–71.4 years of age; oldest, >71.4 years of age. Kaplan-Meier curve per comparison group showing the adjusted cumulative hazards for fracture using follow-up time as timescale (C). Cox proportional hazard model: ICD vs. no diabetes HR 1.47 (95% CI 1.12–1.92), P = 0.005; ACD vs. no diabetes HR 0.91 (0.67–1.23), P = 0.54. Cumulative HR adjusted for femoral neck BMD, age, sex, height, and weight. Light gray, ND; dark gray or dashed, ACD; black, ICD.

Figure 2

Cartoon depicting the differences in bone geometry across glucose control groups for a cross-section of the femoral neck. Individuals with ICD have thicker cortices and narrower neck width than those without diabetes and ACD. With lower instability of cortical bone (lower buckling ratios), the accumulation of microcracks and cortical porosity becomes a possibility to explain bone fragility and fracture susceptibility. Drawing is not to scale.