Factors Associated With the Absence of Cervical Spine Instability in Rheumatoid Arthritis: A >10-Year Prospective Multicenter Cohort Study

Objective: To identify factors associated with the absence of cervical spine instability in patients with rheumatoid arthritis (RA). Methods: Cervical spine instability was defined as the presence of at least one of the following: atlantoaxial subluxation, vertical subluxation of the axis, or subaxial subluxation. In 2001–2002, 634 enrolled outpatients with “classical” or “definite” RA underwent a radiographic cervical spine checkup. In 2012–2013, 233 (36.8%) prospectively underwent routine clinical follow-ups with a >10-year radiographic evaluation. The prevalence and independent predictive factors for no instability were analyzed by multivariable logistic regression. Next, 85 of 292 outpatients (29.1%) without baseline cervical spine instability completed consecutive >5-year and >10-year radiographic examinations. The incidence and predictors for no new development of instability were assessed similarly. Results: Among 233 patients, those without cervical spine instability decreased from 114 (48.9%) to 47 (20.2%) during >10 years. Steinbrocker peripheral joint destruction stages I–II (odds ratio [OR], 3.797; p=0.001), no corticosteroid administration (OR, 2.700; p=0.007), and no previous joint surgery (OR, 2.480; p=0.020) were predictors for no instability. Then, 33 of 85 (38.8%) consecutively followed patients without baseline cervical spine lesions did not develop instability throughout. Steinbrocker stages I–II (OR, 5.355; p=0.005) and no corticosteroid therapy (OR, 3.868; p=0.010) were predictors for no new onset of instability. C-reactive protein (CRP) level≤1.0 mg/dL was marginal in both models (n=233 [OR, 2.013; p=0.057], n=85 [OR, 2.453; p=0.075]). Conclusion: Steinbrocker stages I–II, no corticosteroids, no previous joint surgery, and possibly CRP ≤1.0 mg/dL are factors associated with >10-year absence of cervical spine instability in RA.