Evaluation of bone health and fracture risk in type 2 diabetes: a network meta-analysis of anti-diabetic treatments versus placebo.

Abstract

While studies have highlighted the negative effects of certain antidiabetic agents, similar evidence for other antidiabetic agents remains limited. In this study, we aimed to analyze bone fractures and bone mineral densities (BMDs) across patients receiving antidiabetic treatments, considering both overall and specific sites. We comprehensively searched PubMed, Embase, and ClinicalTrials.gov up to March 2024 to determine the effect of antidiabetic agent use on bone health in patients with type 2 diabetes mellitus. The primary outcome was to reveal variations in fractures across different anti-diabetic treatment modalities. The secondary outcome was the differences in BMDs based on treatment type. The fractures were grouped based on the division of specific sites. We also performed a subgroup analysis to identify differences between treatment types by dividing the study by treatment duration. The protocol is registered (CRD42024538789). A total of 234,759 individuals were enrolled in the 242 studies. We observed a trend wherein all anti-diabetic treatments were associated with decreased risk of fracture compared with placebo; however, this was not significant in direct analysis (OR 0.92, 95% CI 0.84-1.01, I² = 0, P = 0.07). For indirect analysis, glucagon-like peptide-1 receptor agonists (GLP1RAs) demonstrated a significant effect on preventing fractures in non-vertebral fracture, hip fracture, vertebral and hip fracture, and overall fracture (odds ratio [OR] 0.62, 95% confidence interval [CI] 0.43-0.90; OR 0.67, 95% CI 0.49-0.92; OR 0.64, 95% CI 0.47-0.87; OR 0.58, 95% CI 0.48-0.69, respectively). Additionally, dipeptidyl peptidase-4 inhibitors (DPP4i) significantly reduced incidences of non-vertebral fracture, vertebral and hip fracture, and overall fracture risk (OR 0.34, 95% CI 0.25-0.45; OR 0.72, 95% CI 0.55-0.95; OR 0.67, 95% CI 0.55-0.82, respectively). Meanwhile, metformin also reduces overall fracture risk (OR 0.60, 95% CI 0.42-0.88). We observed no significant differences between the antidiabetic agents according to the specific fracture site or study time point. Most antidiabetic treatments except thiazolidinediones did not increase the risk of fractures compared with the placebo. Incretin-based therapies (GLP1RA and DPP4i) exerted beneficial effects on fracture prevention compared with other treatments. These findings underscore the need for well-conducted RCTs to provide further evidence.