Long-Term Drug Therapy and Drug Discontinuations and Holidays for Osteoporosis Fracture Prevention: A Systematic Review

Abstract

Background: Optimal long-term osteoporosis drug treatment (ODT) is uncertain.

Purpose: To summarize the effects of long-term ODT and ODT discontinuation and holidays.

Data sources: Electronic bibliographic databases (January 1995 to October 2018) and systematic review bibliographies.

Study selection: 48 studies that enrolled men or postmenopausal women aged 50 years or older who were being investigated or treated for fracture prevention, compared long-term ODT (>3 years) versus control or ODT continuation versus discontinuation, reported incident fractures (for trials) or harms (for trials and observational studies), and had low or medium risk of bias (ROB).

Data extraction: Two reviewers independently rated ROB and strength of evidence (SOE). One extracted data; another verified accuracy.

Data synthesis: Thirty-five trials (9 unique studies) and 13 observational studies (11 unique studies) had low or medium ROB. In women with osteoporosis, 4 years of alendronate reduced clinical fractures (hazard ratio [HR], 0.64 [95% CI, 0.50 to 0.82]) and radiographic vertebral fractures (both moderate SOE), whereas 4 years of raloxifene reduced vertebral but not nonvertebral fractures. In women with osteopenia or osteoporosis, 6 years of zoledronic acid reduced clinical fractures (HR, 0.73 [CI, 0.60 to 0.90]), including nonvertebral fractures (high SOE) and clinical vertebral fractures (moderate SOE). Long-term bisphosphonates increased risk for 2 rare harms: atypical femoral fractures (low SOE) and osteonecrosis of the jaw (mostly low SOE). In women with unspecified osteoporosis status, 5 to 7 years of hormone therapy reduced clinical fractures (high SOE), including hip fractures (moderate SOE), but increased serious harms. After 3 to 5 years of treatment, bisphosphonate continuation versus discontinuation reduced radiographic vertebral fractures (zoledronic acid; low SOE) and clinical vertebral fractures (alendronate; moderate SOE) but not nonvertebral fractures (low SOE).

Limitation: No trials studied men, clinical fracture data were sparse, methods for estimating harms were heterogeneous, and no trials compared sequential treatments or different durations of drug holidays.

Conclusion: Long-term alendronate and zoledronic acid therapies reduce fracture risk in women with osteoporosis. Long-term bisphosphonate treatment may increase risk for rare adverse events, and continuing treatment beyond 3 to 5 years may reduce risk for vertebral fractures. Long-term hormone therapy reduces hip fracture risks but has serious harms.

Primary funding source: National Institutes of Health and Agency for Healthcare Research and Quality. (PROSPERO: CRD42018087006).