Background Instability of the upper cervical spine is common in patients with rheumatoid arthritis (RA) and can lead to sudden death in worst case scenarios.
Objectives The aim of this study was to evaluate the relationship between upper cervical spine instability and clinical factors, especially focusing on disease activity of RA.
Methods The study included patients who had been enrolled in the Institute of Rheumatology, Rheumatoid Arthritis (IORRA) cohort study within two years from the onset of RA symptoms and who had cervical spine radiographs taken as part of the preoperative testing for elective joint surgery (e.g. joint replacement surgery) at the institute between April 2010 and March 2013. The anterior atlantodental interval (ADI) and Redlund-Johnell (R-J) values were measured to diagnose atlantoaxial subluxation (AAS) and vertical subluxation (VS), respectively. We obtained clinical data from the IORRA cohort study collecting various clinical data biannually, starting in October 2000.
Results Finally, the current study included 56 patients. Nearly half of the patients had AAS (27 patients, 48.2%) and 12 patients (21.4%) had VS. In total, 59.6% of the patients had upper cervical instability. There was no significant correlation between the ADI and R-J values (ρ= -0.025; P=0.8). Multiple linear regression analysis revealed the maximum Disease Activity Score in 28 joints (DAS28) during the study period and height were significantly related to R-J values. No variable was significantly associated with the ADI. Univariate logistic regression analyses indicated only the maximum DAS28 during the study period was significantly associated with developing VS, and no variables were shown to be significant in relation to the development of AAS. Median (IQR) values for the maximum DAS28 during the study period in patients with and without VS were 6.1 (5.0 to 6.6) and 5.1 (4.6 to 5.5), respectively. The area under the ROC curve (AUC) showed the predictions using the maximum DAS28 (cutoff = 5.56, specificity = 79.1%, sensitivity = 69.2%, and AUC =0.714) were moderately accurate (Figure 1).
Conclusions Higher peak disease activity was implicated as a risk factor in the development of VS in RA patients. We suggest that tight control of disease activity is desirable to reduce the risk of VS.
Disclosure of Interest None declared