European Journal of Endocrinology

Effectiveness of anti-osteoporotic medications in the management of glucocorticoid-induced osteoporosis: a systematic review and network meta-analysis on behalf of the ECTS clinical action group

Abstract

OBJECTIVE: To evaluate and compare the efficacy of anti-osteoporotic-medications (AOM) for the prevention and treatment of glucocorticoid-induced-osteoporosis (GIOP). DESIGN: Systematic review and network-meta-analysis conducted in accordance with PRISMA-NMA guidelines (ProsperoID: CRD42024502298). METHODS: MEDLINE, Embase, and Web of Science were searched from inception to December 2025 to identify randomized-controlled-trials and comparative-observational-studies evaluating AOM in adults receiving oral glucocorticoids (GC). Outcomes included percentage change in areal bone- mineral-density (aBMD) at the lumbar-spine (primary), femoral-neck, and total-hip. A random effects network-meta-analysis was performed, integrating direct and indirect comparisons across AOM. Certainty of evidence was assessed using the CINeMA framework. RESULTS: Thirty-one studies (28 randomized-controlled-trials and 3 observational-studies) involving 5260 participants (age-range: 29.6-70.6 years, 72.5% females) were included. All AOM were superior to placebo or calcium/vitaminD supplementation in increasing lumbar-spine aBMD. Teriparatide demonstrated the greatest efficacy compared with alendronate (effect-size: 3.8; 95% CI: 2.9-4.6), risedronate (4.0; 95% CI: 2.9-5.1), denosumab (1.5; 95% CI: 0.1-2.9), and zoledronate (2.4; 95% CI: 1.1-3.6), while no significant difference was observed with romosozumab (0.3; 95% CI: -2.1 to 1.4). Among antiresorptive-medications, zoledronate and denosumab showed greater efficacy than oral bisphosphonates, which demonstrated modest positive effects. In sensitivity analyses restricted to established GIOP, defined as GC users ≥3months, teriparatide and romosozumab remained superior to antiresorptive-medications. Evidence for total-hip and femoral-neck aBMD outcomes was limited by substantial network inconsistency (mostly missing data). Included studies were underpowered to assess difference in fracture outcomes. CONCLUSIONS: This network-meta-analysis demonstrates a clear gradient of efficacy among AOM in GIOP, with anabolic and potent antiresorptive-medications producing the largest gains in lumbar-spine aBMD supporting current ECTS recommendations advocating risk-stratified treatment selection.