Background/Purpose: Women and men with rheumatoid arthritis [RA] are at increased risk for a fragility fracture [fx]. To help better target those in need of fx prevention, we sought to determine which RA characteristics were associated with this increased risk for fragility fxs and specifically examined the effect of chronic glucocorticoid [GC] use.

Methods: We studied a population-based inception cohort (age ≥18 yrs), who fulfilled 1987 ACR criteria for RA between 1955-2007, and were followed until death or last available follow-up. RA characteristics (presence of rheumatoid factor [RF+], erosions, nodules, and other extra-articular manifestations), major joint surgeries, oral GC use, and all incident fxs after RA diagnosis were identified through a review of complete (inpatient and outpatient) community medical records. We studied fragility fxs, which excluded pathologic fxs and fxs resulting from severe trauma (e.g. motor vehicle accident). Chronic GC use was defined as >3 months at any dose. Using age-and sex-adjusted Andersen-Gill models, we examined the impact of RA characteristics on the subsequent risk for fxs. Most risk factors were modeled as time-dependent covariates. We performed similar analyses stratified by sex and then chronic GC use.

Results: In 1171 RA subjects, (822 women [66% RF+] and 349 men [69% RF+]; mean [± SD] age at RA diagnosis: 56 ± 16 years and 58 ± 14 years, respectively), followed for 15,195 person-years, there were 828 fragility fxs in 328 women and 253 fragility fxs in 112 men since their RA diagnosis.  Over follow-up, about 1/3 of women and men each had nodules; a similar proportion developed erosions; ~12% each had other extra-articular manifestations; and 29% underwent major joint surgery. There were 424 (52%) women and 209 (60%) men who were chronic GC users; the remainder either never used GC for RA (70%) or had only limited GC exposure.  We found that RF+, erosions, nodules, extra-articular manifestations, major joint surgeries and chronic GC use were all associated with an increased risk for fragility fxs (see Table). Extra-articular manifestations and major joint surgeries were associated with an increased risk for fxs regardless of GC exposure (data not shown). However, the other RA characteristics tended to be associated with increased fx risk only in chronic CG users, but not in those with no/limited GC use (data not shown).

Conclusion: RA characteristics that are often associated with greater disease severity are also associated with increased fx risk. However, only extra-articular manifestations and major joint surgeries were associated with increased fx risk regardless of chronic GC use. Although further work is required to understand their interaction on fx risk, some RA characteristics may simply be indicators of chronic GC use; on the other hand, others may either better represent greater disease activity or reflect greater frailty and/or fall risk.