Background/Purpose: Osteoporotic (OP) fractures are a major cause of disability, cost, and mortality in RA. Besides increased OP fracture risk, chronic inflammation and pain predispose RA patients to several comorbidities including cardiovascular, mental, and gastrointestinal disorders that lead to the frequent use of multiple medications. Some of these medications have been reported to influence OP fracture risk in the general population, but these associations have not been studied in RA patients. We examined the association of DMARDs, statins, antidepressants, proton pump inhibitors (PPI), opioids, NSAIDs, anticonvulsants, and antipsychotics with OP fracture risk in a US-wide observational RA cohort.

Methods: RA patients (≥40 years old) without prior OP fracture from 2001 through 2016 in the National Data Bank for Rheumatic Diseases (NDB) were assessed for OP fractures (fractures resulting from any fall from a standing height or less). DMARDs were categorized into 4 mutually exclusive groups: (1) MTX monotherapy-reference (2) TNF inhibitors (TNFi) (3) Non-TNFi biologics (4) Others; along with a separate glucocorticoid (GC) variable. Statins, antidepressants (selective serotonin reuptake inhibitors [SSRI] and others), PPI, opioids (weak and strong opioids), NSAIDs, anticonvulsants, and antipsychotics were evaluated separately as current use vs. nonuse and based on the treatment duration. Cox proportional hazard models were used to adjust for sociodemographics, comorbidities, BMI, fracture risk by FRAX, and RA severity measures.

Results: During a median (IQR) 5.4 (2.3-9.7) years of followup in 11,002 RA patients, 863 (7.8%) OP fractures were observed. Patients who developed fractures were significantly older, had higher disease duration and activity, GC use, comorbidity and FRAX scores at baseline than who did not. Crude incidence rates (95% CI) in each medication group are presented in the Table. The adjusted models showed a significant OP fracture risk increase with GC use of ≥3 months of any dose, SSRI (HR 1.35 [1.10-1.64], P=0.003), opioids of any strength (weak: HR 1.48 [1.26-1.74]; strong: HR 1.78 [1.41-2.26], P<0.001 for both) (Table). OP fracture risk increase started even after 1-30 days of opioid use (HR 1.99 [1.59-2.48], P<0.001), whereas SSRI-associated risk increase started after 3 months of use (HR 1.25 [0.99-1.59], P= 0.061).

Conclusion: SSRI and opioids were associated with increased OP fracture risk in RA patients. This risk increase might be associated with increased fall risk caused by these medications. Given the frequent occurrence of chronic pain, mood disorders and polypharmacy in RA patients, the necessity of the medications should regularly be reviewed in RA patients, particularly who have high fracture risk and long-term GC use. When managing pain with opioids, even in the short-term, clinicians should be aware of the fracture risk.