Purpose: To evaluate the effect of different non-osteoporotic drugs on the increase or decrease in the risk of incident fragility fractures (vertebral, humerus or hip) in a cohort of patients diagnosed with osteoporosis on active anti-osteoporotic therapy.
Methods: For this retrospective longitudinal study, baseline and follow-up data on prescribed non-osteoporotic treatments and the occurrence of vertebral, humerus or hip fractures in 993 patients from the OSTEOMED registry were analyzed using logistic regression models. The drugs evaluated with a possible beneficial effect were thiazides and statins, while the drugs evaluated with a possible harmful effect were antiandrogens, aromatase inhibitors, proton pump inhibitors, selective serotonin reuptake inhibitors, benzodiazepines, GnRH agonists, thyroid hormones, and oral and inhaled corticosteroids.
Results: Logistic regression analyses indicated that no treatment significantly improved fracture risk, with the only treatments that significantly worsened fracture risk being letrozole (OR = 0.18, p-value = 0.03) and oral corticosteroids at doses ≤ 5 mg/day (OR = 0.16, p-value = 0.03) and > 5 mg/day (OR = 0.27, p-value = 0.04).
Conclusion: The potential beneficial or detrimental effects of the different drugs evaluated on fracture risk are masked by treatment with anabolic or antiresorptive drugs that have a more potent action on bone metabolism, with two exceptions: letrozole and oral corticosteroids. These findings may have important clinical implications, as patients receiving these treatments are not fully protected by bisphosphonates, which may imply the need for more potent anti-osteoporotic drugs such as denosumab or teriparatide.
Keywords: Cohort analysis; Fractures; Logistic models; Osteoporosis; Prescription drugs.
© 2023. The Author(s).