Prevalence and risk factors of anterior atlantoaxial subluxation in ankylosing spondylitis

Abstract

Objective. In ankylosing spondylitis (AS), the cervical spine, like other sections of the spine and sacroiliac joints, is vulnerable during the disease process. Atlantoaxial subluxation (AAS) has been studied in connection with AS, but its risk factors and progression have not been clarified. Therefore, this study assessed the prevalence and risk factors of AAS in patients with AS.Methods. A total of 819 patients with AS who fulfilled the modified New York criteria and were examined with a full-flexion lateral view of the cervical spine by radiograph were enrolled from an outpatient clinic. The medical records of the patients were retrospectively reviewed and the anterior atlantodental interval (AADI) in the lateral flexion view of the cervical spine radiograph was investigated by 2 experienced musculoskeletal radiologists. We defined the AAS as an AADI of > 3 mm, and progression of AADI as a progression rate > 0.5 mm/year.Results. AAS was found in 14.1% (116/819) of patients. Progression of AADI occurred in 32.1% (26/81) patients with AAS and 5.0% (16/320) patients without AAS (p < 0.001). The development of AAS was significantly associated with elevated C-reactive protein [CRP; OR 2.19 (1.36-3.53)], peripheral arthritis [OR 2.05 (1.36-3.07)], use of anti-tumor necrosis factor antagonists because of failure of nonsteroidal antiinflammatory drugs/disease-modifying antirheumatic drugs [NSAID/DMARD; OR 2.28 (1.52-3.42)], and uveitis [OR 1.71 (1.13-2.59)]. These factors were adjusted for age, sex, and disease duration by logistic regression analysis. No clear association was found for HLA-B27, seropositivity, or smoking status with AAS.Conclusion. AAS is a frequent complication, and the progression of AADI was more rapid in cases with AAS. The presence of peripheral arthritis, or high disease activity with elevated CRP level or refractory to conventional NSAID/DMARD, independently increased the risk of AAS, suggesting that clinicians should focus on the detection and monitoring of AAS, especially in cases with associated risk factors. (First Release Oct 1 2012: J Rheumatol 2012;39:2321-6; doi:10.3899/jrheum.120260)