Bisphosphonate Use and the Risk of Undergoing Total Knee Arthroplasty in Osteoporotic Patients with Osteoarthritis

BACKGROUND: The use of bisphosphonates has been reported to have potential beneficial effects on knee osteoarthritis, but existing studies have limitations. The purpose of this study was to examine the association of bisphosphonate use with the risk of undergoing total knee arthroplasty and with the consumption of pain medication among osteoporotic patients with knee osteoarthritis. METHODS: We identified patients who were newly diagnosed with knee osteoarthritis among a cohort of patients with osteoporosis from 2009 to 2012 in the National Health Insurance Research Database in Taiwan. We further categorized these patients into 2 groups: those who were treated with bisphosphonates (bisphosphonate users) and those who were not treated with any anti-osteoporosis drug (nonusers). Bisphosphonate treatment adherence was calculated by the medication possession ratio (MPR) as the proportion of days of bisphosphonate treatment within a fixed duration; an MPR of ≥80% was considered high adherence. The primary and secondary outcomes of interest were undergoing total knee arthroplasty and the use of pain medication, respectively. Analyses using Cox proportional hazard models with propensity-score adjustment were performed to estimate the association between bisphosphonate use and the risk of undergoing total knee arthroplasty. The incremental change in the mean accumulated defined daily doses of pain medications among both bisphosphonate users and nonusers was calculated. RESULTS: We identified 16,276 bisphosphonate users and 123,791 nonusers of any anti-osteoporosis drug among the patients with osteoporosis who were newly diagnosed with osteoarthritis. Bisphosphonate use was significantly associated with a decreased risk of total knee arthroplasty (adjusted hazard ratio HR, 0.76; 95% confidence interval CI, 0.69 to 0.83; p < 0.001). In patients with a follow-up period of ≥24 months and an MPR of ≥80%, the effect size was significantly greater (adjusted HR, 0.66; p = 0.048). Over the 5 years of follow-up, we found a significantly greater decrease in the use of pain medication among bisphosphonate users than among nonusers (p < 0.001; Chow test). CONCLUSIONS: Among patients with osteoporosis and osteoarthritis, bisphosphonate use was associated with a significantly lower risk of total knee arthroplasty, especially in patients with high adherence and longer treatment duration. A lower consumption of pain medication was also found for bisphosphonate users among the patients with osteoporosis and osteoarthritis. LEVEL OF EVIDENCE: Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.