Glucocorticoid-induced osteoporosis: is the bone density decrease the only explanation?

Abstract

BACKGROUND: Glucocorticoids may increase bone fragility via mechanisms independent from their bone mass reducing effect. OBJECTIVE: To study relationships between osteoporotic fractures and bone mineral density in patients on long-term glucocorticoid therapy. PATIENTS AND METHODS: We studied 121 women with a mean age of 60.4 +/- 14.3 years on long-term glucocorticoid therapy (cumulative dose > or = 1 g of prednisone equivalent, duration > or = 6 months) for rheumatoid arthritis (n = 38), polymyalgia rheumatica or giant cell arteritis (n = 26), connective tissue disease (n = 15), asthma (n = 14), another inflammatory joint disease (n = 14), or another condition (n = 14). The control group was composed of 125 subjects who had the same mean age and met the same exclusion criteria as the case group. Bone mineral density was measured at the lumbar spine and femoral neck using a Hologic QDR 4500 unit. In subjects with back pain, radiographs of the thoracic and lumbar spine were obtained to look for fractures. RESULTS: The odds ratio for a bone mineral density decrease of one standard deviation at the femoral neck was 1.68 (1.20-2.35) in patients with a cumulative glucocorticoid dose of 10 g of prednisone equivalent and 1.67 (1.22-2.29) in those with a glucocorticoid therapy duration of 2 years. Sixty-eight fractures were recorded in 56 patients (46% of the overall patient group). Even after adjustment on age, glucocorticoid therapy duration, and dose, mean bone mineral density values at the lumbar spine and femoral neck were significantly lower in the subgroup of patients with fractures than in the subgroup without fractures. Sensitivity and specificity of bone mineral density at the femoral neck and/or lumbar spine for the diagnosis of vertebral fracture and/or peripheral fracture were 73% and 51%, respectively. In the stepwise logistic regression model, factors explaining the presence of fractures were as follows, in hierarchical order: age; absence of calcium/vitamin D supplementation, femoral neck T-score, and glucocorticoid dose. CONCLUSION: Our data are compelling evidence that bone mineral density is a major determinant of the fracture risk in patients with glucocorticoid-induced osteoporosis.