Metformin and Lifestyle Interventions in DPP Do Not Have Long-Term Effects on Bone Density or Falls

Abstract

Randomization to lifestyle intervention (ILS) or metformin (MET) in the Diabetes Prevention Program (DPP) resulted in weight loss and reduced progression to diabetes. The long term effects of these interventions on bone mineral density (BMD) and falls, both associated with weight loss, are not known. We determined if ILS or MET was associated with prevalence of osteoporosis, osteopenia or history of falls in the DPP Outcome Study (DPPOS). During DPP, 3234 participants at high risk for diabetes were randomized to ILS, MET or PLB; 2779 continued in DPPOS and were offered modified group ILS. During DPPOS, ILS was offered additional lifestyle reinforcement semi-annually and MET received unmasked metformin. This analysis includes 1384 DPPOS participants who had dual-energy X-ray absorptiometry hip scans at year 12 (Y12). Osteoporosis was defined as femoral neck BMD T-score ≤-2.5 and osteopenia as -2.5<T-score≤-1. Participants were asked at Y12 if they had a fall in the previous year. Outcomes were compared across the 3 treatment groups using Pearson’s chi- squared test. Associations between treatment and outcomes were assessed with logistic regression. Mean age at Y12 was 66.5 (±9.5) years. Mean time since DPP baseline was 15.6 (±0.7) years. Weight loss (DPP baseline to DPPOS Y12) did not differ across treatment groups (-4.4 kg ILS; -3.5 kg MET; -3.1 kg PLB; p=0.114). There were no differences across groups in the prevalence of osteoporosis in men (1.4% ILS, 1.3% MET, 0.7% PLB; p=0.836) or women (2.9% ILS, 2.9% MET, 3.1% PLB; p=0.983). Prevalence of osteopenia was also similar across groups. Prevalence of history of falls was similar across groups in men (16.0% ILS, 16.4% MET, 16.0% PLB; p=0.995) and women (25.7% ILS, 21.3% MET, 21.6% PLB; p=0.342). Prevalence of osteoporosis, osteopenia or history of falls did not differ between ILS or MET and PLB in men or women. In a cohort at high risk of diabetes, lifestyle intervention or metformin did not have long-term effects on bone density or falls. Disclosure A.V. Schwartz: None. Q. Pan: None. V. Aroda: Consultant; Self; Novo Nordisk Inc.. Research Support; Self; Novo Nordisk Inc., AstraZeneca, Calibra Medical, Eisai Inc., Sanofi. Consultant; Self; Sanofi. Research Support; Self; Theracos, Inc.. Employee; Spouse/Partner; Merck & Co., Inc.. Other Relationship; Self; American Diabetes Association. Consultant; Self; ADOCIA. J.P. Crandall: None. A. Kriska: None. C.S. Piromalli: None. M. Temprosa: Other Relationship; Self; Merck KGaA. H. Florez: Advisory Panel; Self; Sanofi. D. Research Group: None.