Prolonged glucocorticoid use and rheumatoid arthritis-associated cervical spine deformity

Objectives RA patients are often prescribed glucocorticoids, although it is known that their long-term use increases the risk of osteoporosis and fractures. The association between glucocorticoid use and RA-associated cervical spine deformity is yet to be determined. Method Duration and dose of glucocorticoid use were evaluated in patients with new onset RA (BeSt Trial). Missing values on the exposure were imputed using the last observation carried forward. Lateral X-rays at 5- and 10-year follow-ups were assessed for atlantoaxial subluxation (AAS) and subaxial subluxation (SAS). To estimate the association between glucocorticoids and cervical spine deformity, multiple logistic regression models adjusted for age, gender, baseline Disease Activity Score (DAS), ACPA positivity, and RF positivity were used to estimate odds ratios (ORs) and their 95% confidence intervals (CIs). Mediation analysis was performed to evaluate whether such potential association was mediated via mean DAS. Results Cervical deformity (AAS and/or SAS > 2 mm) was observed in 108 (40%) out of 272 patients. For a 1-year increase in total duration of glucocorticoid use, the adjusted OR for RA-associated cervical spine deformity was 1.19 (95% CI, 1.03-1.38; p = 0.02), and for an increase of 1 g of glucocorticoid in total cumulative dose, the OR was 1.06 (95% CI, 1.01-1.12; p = 0.02). Mediation analysis could not reveal an influence of mean DAS on these associations. Conclusions There was evidence of a direct association between long-term use of glucocorticoids in newly diagnosed RA patients and RA-associated cervical spine deformity after 10 years. Other effective therapies to suppress disease activity may be preferred over glucocorticoids.