Retrospective data analysis.Objective.
To identify factors affecting the atlantodental interval, the Ranawat value, and subaxial translation after rheumatoid arthritis (RA) diagnosis. In addition, factors predictive for cervical spine instability (CSI) development after RA diagnosis were examined.Summary of Background Data.
Development of CSI affects the prognosis and mortality of RA patients. Previous studies described that obesity is associated with reduced radiographic joint damage in RA patients. We hypothesized that body mass index (BMI) is also associated with radiographic cervical damage in RA patients.Methods.
Cervical radiographs were taken at full flexion, neutral position, and full extension to measure the geometric length of the anterior atlantodental interval, the Ranawat value, and subaxial translation. These values were entered into multivariable linear regression analysis based on potential associated factors. Hazard ratios were calculated to identify independent factors predictive of CSI.Results.
Of the patients diagnosed with RA between January 2005 and August 2015, 1611 who underwent at least one cervical radiograph were included. After adjusting for sex, age, BMI category, CSI, rheumatoid factor, and RA medication, multivariate analysis revealed that the risk of atlantoaxial subluxation in the underweight and normal BMI groups was about 1.6-fold (hazard ratio, 1.63; 95% CI, 1.10–2.43; P = 0.015) and 1.7-fold higher, respectively, than that in the obese group, and that the risk of vertical subluxation was about 2.5-fold (hazard ratio, 2.52; 95% CI, 1.32–4.83; P = 0.005) higher in the underweight group than in the obese group. We also found that the rheumatoid factor positivity was a predictive risk factor for CSI development.Conclusion.
We identified risk factors predictive for CSI occurrence after RA diagnosis through cervical radiograph assessment. We found that BMI was an independent predictor for development of CSI. Further large-scale prospective studies are required to confirm these findings.Conclusion.
Level of Evidence: 3