Single-center retrospective DXA study matching semaglutide and tirzepatide users (≥6 months) with non-users by age, sex, BMI, and diabetes status, measuring bone mineral density changes in a population with elevated fracture risk. Evaluates whether GLP-1/GIP receptor agonism affects bone density beyond weight-loss-associated BMD changes. Provides dedicated clinical data on semaglutide and tirzepatide skeletal effects—an important safety concern given that significant weight loss typically reduces bone mineral density and the growing use of these agents in postmenopausal women and older adults with osteoporosis risk.
Abstract
CONTEXT: Glucagon-like peptide-1 receptor agonists (GLP1-RA) have potent glucose-lowering and weight loss benefits, but their effects on bone remain unclear.
OBJECTIVE: To investigate changes in bone mineral density (BMD) in patients using semaglutide (SEM) and tirzepatide (TIR), a dual agonist of GLP-1/glucose dependent insulinotropic polypeptide.
METHODS: Single-center retrospective study. Adult patients using SEM/TIR for ≥6 months with DXA scans before initiation and at least 6 months after were matched by age, sex, BMI, and diabetes mellitus (DM) to non-users with at least two DXA scans over the same period. The primary outcome was percentage change in total hip (TH) BMD.
RESULTS: We included 255 patients using SEM or TIR in the GLP-1 RA group (92% female, mean age 64±9 years, BMI 31.0±5.6 kg/m²) and 255 controls. After a median follow-up of 17 months, the GLP-1 RA group achieved median 5% weight loss. Both groups had significant declines in BMD at TH and FN, with similar magnitude between groups. In the GLP-1RA group, weight loss was directly associated with bone loss at the TH and FN (r=0.32 for TH, r=0.17 for FN, both p<0.01). Among patients without DM, greater TH bone loss was noted in GLP-1 RA group compared to controls (-1% vs -0.6%, p=0.04), whereas TH bone loss was similar between groups among patients with DM.
CONCLUSION: SEM/TIR use was associated with greater annualized TH bone loss in patients without DM, whereas TH bone loss was comparable between GLP-1 RA and controls in patients with DM. These findings suggest GLP-1 RA's effects on bone may differ by DM status, with weight loss driving bone loss in patients without DM.