Background/Purpose: Women and men with RA are at an increased risk for fracture [fx] as well as greater overall mortality. It is recognized that in the general population, overall mortality among those with a fx is increased. We sought to determine whether an incident fx following the diagnosis of RA contributes to an excess mortality beyond what would be expected for an incident fracture in those without RA.

Methods: We studied a population-based inception cohort (age ≥18 yrs), who fulfilled 1987 American College of Rheumatology criteria for RA between 1955-2007, and an equal number of age- and sex-matched non-RA subjects from the same underlying population, who were followed until death or last available follow-up. All incident fxs were identified through a complete review (inpatient and outpatient) of medical records. We excluded from analyses pathologic fxs and fxs resulting from severe trauma (e.g. motor vehicle accidents, etc.). Cox proportional models were used to examine the risk for death following the first incident non-pathologic fracture resulting from no more than moderate trauma [moderate trauma fx] after RA diagnosis, or equivalent index date in non-RA subjects. All analyses were adjusted for age and sex, and fractures were modeled as a time-dependent covariate. In RA subjects, we also examined in separate analyses whether steroid use, being seropositive, or having nodules or erosions (modeled as time-dependent covariates) could account for the risk of death following fracture.

Results: In 1171 RA subjects, (822 women and 349 men, mean age at RA diagnosis ± SD: 56 ± 16 yrs and 58 ± 14 yrs, respectively), followed for 15,707 person-yrs [p-y], 440 had any moderate trauma fx, while 535 died over the available follow-up. In sex- and age-matched non-RA subjects, 374 had any moderate trauma fx and 441 died over 17,643 p-y follow-up. In RA subjects, a moderate trauma fx increased the risk for death (Hazard Ratio [HR], 2.2; 95% CI: 1.8, 2.6). An increased risk for death was also observed in non-RA subjects following a moderate trauma fx (HR: 1.5, 95% CI: 1.2, 1.9) although this risk was significantly lower (p = 0.04) than in RA. In RA subjects, any moderate trauma fracture remained a significant predictor of death even after adjusting for steroid use, being seropositive, or having nodules or erosions (HR: 1.9, 95% CI: 1.6, 2.3).

Conclusion: A moderate trauma fracture in RA is associated with a greater than expected excess mortality.  Identifying characteristics or other co-morbidities among those with RA who fracture may help explain this excess risk of death following fractures.

---

« Back to 2012 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/greater-than-expected-increased-mortality-following-fragility-fractures-in-women-and-men-with-rheumatoid-arthritis/