Objective: Increased bone turnover is associated with SGLT2i use. Patients with diabetes have adverse effects on bone metabolism. Our aim was to evaluate if SGLT2i was associated with fractures versus DPP4i.
Methods: A retrospective cohort combined Veterans Administration, Medicare, and National Death Index databases. New users of SGLT2i with diabetes (87% empagliflozin; 10.2% canagliflozin; 2.6% dapagliflozin) or DPP4i (51% saxagliptin; 30% sitagliptin; 12% alogliptin) were followed from the prescription fill until a fracture event, death or study end. Fractures included: face/skull, spine, ribs, long bones, hand/feet/ digits, or hip. Fractures were identified based on a validated algorithm with PPV 91.3% (86.8, 94.4) compared to medical records. Cox models compared fractures between SGLT2i and DPP4i in a propensity score weighted cohort adjusted for clinical and lab data including vitamin D levels, smoking, and medications.
Results: The weighted cohort included 36253 SGLT2i vs 36189 DPP4i episodes. Median age was 69 years and diabetes duration 9.3 (5.7, 13.7) years. In matched weighted analyses, there were 468 and 549 fractures among SGLT2i and DPP4i users, respectively. There were no clinical differences in fractures per 1000 person years: 18.1 (16.6, 19.7) vs. 20.2 (19.1, 21.4) [aHR 0.89 (0.80, 0.99)].
Conclusions: SGLT2i use for diabetes was not associated with increased fracture outcomes compared to DPP4i
K.D. Snyder: None. A.J. Hackstadt: None. K.E. Griffin: None. T.G. Horton: None. A. Javid: None. A.M. Hung: None. R.A. Greevy: None. C. Roumie: None.
VA Clinical Science research and Development investigator initiated grant CX000570-11 (Roumie)
