Chronic administration of glucocorticoids is associated with bone loss and increased fracture risk, particularly in the trabecular bone of the vertebrae.OBJECTIVE
To assess the prevalence of asymptomatic vertebral fractures in postmenopausal women undergoing long-term glucocorticoid therapy.DESIGN AND INTERVENTION
This was a multicenter, cross-sectional outpatient study of postmenopausal women. Inclusion criteria included age >45 years, glucocorticoid therapy for ≥6 months with a cumulative dose ≥1.35 g of a prednisone equivalent, and diagnosis of rheumatoid arthritis, systemic lupus erythematosus, polymyalgia rheumatica, other vasculitides or connective tissue diseases, asthma, or chronic obstructive pulmonary disease. Patients with back pain, previous vertebral fracture, Paget's disease, or using antiosteoporotic medication were excluded; however, calcium and vitamin D supplements were permitted. Participants underwent a medical interview that assessed socio-demographic factors, details of glucocorticoid therapy, and risk factors for osteoporosis. Health-related quality of life was evaluated with the European Foundation for Osteoporosis questionnaire. A back-function questionnaire rated the participants' ability to perform activities of daily living. Vertebral fractures were identified from lateral and antero-posterior thoracolumbar radiographs. Hip and lumbar spine BMD were assessed by dual energy X-ray absorptiometry and quantitative ultrasound.OUTCOME MEASURE
The primary outcome measure was the prevalence of asymptomatic vertebral fractures.RESULTS
The study enrolled a total of 551 women (median age 65.2 years). Rheumatoid arthritis was the most common reason for glucocorticoid therapy (53.2%). One or more asymptomatic vertebral fractures were present in 37.02% of participants, whereas 14.42% had two or more fractures. A bimodal distribution of these fractures was observed, with peaks at the T7 and T11 thoracic vertebrae. The prevalence of asymptomatic vertebral fractures increased with age; 47.98% of participants ≥70 years of age presented with at least one fracture (P = 0.006). There was a strong association between polymyalgia rheumatica and fracture prevalence; 51.09% of affected patients had one or more fracture (P = 0.012). When the data were adjusted for age, cumulative glucocorticoid dose, therapy duration, and fracture history, however, this association became weaker (43.20%), and was similar to the fracture prevalence of 43.36% observed in patients with other vasculitides or connective-tissue diseases. There was no correlation between fracture prevalence and cumulative glucocorticoid dose or therapy duration. Additionally, lumbar spine and hip BMD, calcaneal bone stiffness, and health-related quality of life were not associated with the number or severity of vertebral fractures. Time since menopause and history of nonvertebral fractures both showed a significant association with asymptomatic vertebral fracture (P<0.001 and P = 0.002, respectively). Back function score showed a modest association with the number of asymptomatic vertebral fractures.CONCLUSION
A high prevalence of asymptomatic vertebral fractures was observed in postmenopausal women undergoing long-term glucocorticoid therapy, which suggests a need for careful evaluation of this patient subgroup.