The use of thiazolidinediones (TZDs) has been associated with an increased fracture risk. In addition, type 2 diabetes mellitus (T2DM) has been linked with fracture. We evaluated to what extent the association between TZD use and fracture risk is related to the drug or to the underlying disease. We conducted a population-based cohort study using the Danish National Health Registers (1996–2007), which link pharmacy data to the national hospital registry. Oral antidiabetic users (n = 180,049) were matched 1:3 by year of birth and sex to nonusers. Cox proportional hazards models were used to estimate hazard ratios (HRs) of fracture. Time-dependent adjustments were made for age, comorbidity, and drug use. We created a proxy indicator for the severity of disease. The first stage was defined as current use of either a biguanide or a sulfonyluerum, the second stage as current use of a biguanide and a sulfonyluerum at the same time, the third stage as patients using TZDs, and the fourth stage as patients using insulin. The risk of osteoporotic fracture was increased 1.3-fold for stages 3 and 4 compared with controls. Risk with current TZD use (stage 3 HR = 1.27, 95 % CI 1.06–1.52) and risk with current use of insulin (stage 4 HR = 1.25, 95 % CI 1.20–1.31) were similar. In the first (HR = 1.15, 95 % CI 1.13–1.18) and second (HR = 1.00, 95 % CI 0.96–1.04) stages risks were lower. Risk of osteoporotic fracture was similar for TZD users and insulin users. When studying fracture risk with TZDs, the underlying T2DM should be taken into account.

The pharmacological management of disease should involve consideration of the balance between the beneficial effects of treatment on outcome and the probability of adverse effects. The aim of this review is to explore the risk of adverse drug reactions and drug–drug interactions with treatments for postmenopausal osteoporosis. We reviewed evidence for adverse reactions from regulatory documents, randomized controlled trials, pharmacovigilance surveys, and case series. Bisphosphonates are associated with gastrointestinal effects, musculoskeletal pain, and acute-phase reactions, as well as, very rarely, atrial fibrillation, atypical fracture, delayed fracture healing, osteonecrosis of the jaw, hypersensitivity reactions, and renal impairment. Cutaneous effects and osteonecrosis of the jaw are of concern for denosumab (both very rare), though there are no pharmacovigilance data for this agent yet. The selective estrogen receptor modulators are associated with hot flushes, leg cramps, and, very rarely, venous thromboembolism and stroke. Strontium ranelate has been linked to hypersensitivity reactions and venous thromboembolism (both very rare) and teriparatide with headache, nausea, dizziness, and limb pain. The solidity of the evidence base depends on the frequency of the reaction, and causality is not always easy to establish for the very rare adverse reactions. Drug–drug interactions are rare. Osteoporosis treatments are generally safe and well tolerated, though they are associated with a few very rare serious adverse reactions. While these are a cause for concern, the risk should be weighed against the benefits of treatment itself, i.e., the prevention of osteoporotic fracture.