Time-dependent risk of fracture in adults with type 2 diabetes receiving anti-diabetic drug: A one-stage network meta-analysis

Abstract

Aims: To assess the time-dependent risk of fracture in adults with type 2 diabetes receiving anti-diabetic drugs.

Materials and methods: We searched MEDLINE, EMBASE, and Cochrane Library up to 18 November 2021, for randomized controlled trials (RCTs) and propensity-score-matched non-randomized studies (NRSs) comparing all anti-diabetic drugs with standard treatment or with each other on fracture in adults with type 2 diabetes. The study performed a one-stage network meta-analysis using discrete-time hazard regression with reconstructed individual time-to-event data.

Results: This network meta-analysis involved seven RCTs (65,051 adults with type 2 diabetes) with a median follow-up of 36 months and three propensity-score-based NRSs (17,954 participants) with a median follow-up of 27.3 months. Among anti-diabetic drugs, thiazolidinediones increased the overall hazard of fracture by 42% (95% credible interval [CrI], 3%-97%) and almost tripled the risk after 4 years (hazard ratio [HR], 2.74; 95% CrI, 1.53-4.80). Credible subgroup analysis suggested that thiazolidinediones increased the hazard of fracture only in females (HR, 2.19; 95% CrI, 1.26-3.74) but not among males (HR, 0.81; 95% CrI, 0.45-1.40). Moderate certainty evidence established that thiazolidinediones increase 92 fractures in five years per 1000 female patients. We did not find the risk of fractures with other anti-diabetic drugs including metformin, sulfonylureas, sodium-glucose cotransporter-2 (SGLT2) inhibitors, and dipeptidyl peptidase-4 (DPP-4) inhibitors.

Conclusions: Long-term use of thiazolidinediones elevates the risk of fracture among females with type 2 diabetes. There is no evidence eliciting fracture risk associated with other anti-diabetic drugs.

Keywords: anti-diabetic drug; bone fracture; network meta-analysis; systematic review.