Background/Purpose: Patients with rheumatoid arthritis (RA) have a much greater risk of sustaining a hip fracture than those without RA. Hip fractures cause significant morbidity and mortality. Some studies have reported beneficial effects of tumor necrosis alpha-inhibitors (TNFα-I) on markers of bone health. However, whether use of TNFα-I decreases the risk of hip fractures in patients with RA remains unclear.

Methods: Using national VA data from 2000-2011, a retrospective cohort of adult veterans with RA was identified using an algorithm that combined coded diagnoses with use of specific antirheumatic medications. This algorithm had a positive predictive value of 81% for identification of patients with RA compared with manual chart review. Two years of continuous baseline enrollment was required prior to cohort entry. Veterans with prior hip fracture, cancer, and serious renal, liver, or lung disease identified during the baseline period were excluded. From this RA cohort, cases of incident hip fracture that required hospitalization were identified using a validated algorithm. Incidence density sampling was used to select up to 20 at-risk controls per case. Controls were matched to cases on age, VA integrated service network and time since cohort entry. Pharmacy data were used to assess disease modifying anti-rheumatic drug (DMARD) exposure in the 12 months preceding the index date of hip fracture. To be considered a user of a given medication, a minimum of 180 days of use was required. Cases and controls were classified into one of four mutually exclusive exposure categories: non-biologic DMARD users (reference), TNFα-I biologic DMARD users, non-TNFα-I biologic DMARD users, and other users. Multivariable conditional logistic regression was used to assess the association between medication exposure and the risk of hip fracture by computing odds ratios adjusted for demographics, glucocorticoid use, BMI and other risk factors for hip fractures.

Results: During follow-up, 462 hip fracture cases and 8226 matched controls were identified from the underlying cohort of veterans with RA. The mean age of cases and controls was 75 years. In multivariable analyses that accounted for relevant covariates, hip fracture cases had similar odds of exposure to TNFα-I compared with controls (adjusted odds ratio [aOR]: 0.97 [95% confidence interval, 0.60 – 1.60]).

Conclusion: In this preliminary assessment, use of TNFα-I was not associated with a decreased risk of hip fractures compared with use of non-biologic DMARDs among veterans with RA. The cohort is currently being updated to include data through 2017.

« Back to 2019 ACR/ARP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/tumor-necrosis-factor-alpha-inhibitor-tnf-%ce%b1-inhibitor-exposure-and-risk-of-hip-fracture-in-veterans-with-rheumatoid-arthritis-a-nested-case-control-study/