Multivariate logistic regression analyses, in which baseline fat, inflammation and syndesmophytes/ankylosis were entered into a generalised linear latent and mixed model, to explore their association with the development of new syndesmophytes after 2 years in 76 patients with ankylosing spondylitis
| Variables entered into model | OR (95% CI) | p Value |
|---|---|---|
| Model A | ||
| Baseline type A CIL | 0.91 (0.32 – 2.63) | 0.87 |
| Baseline syndesmophytes/ankylosis | 1.19 (1.11 – 1.27) | <0.0001 |
| Model B | ||
| Baseline type B CIL | 3.88 (1.20 – 12.57) | 0.024 |
| Baseline syndesmophytes/ankylosis | 1.21 (1.12 – 1.31) | <0.0001 |
| Model C | ||
| Baseline fat lesion | 4.83 (2.38 – 9.80) | <0.0001 |
| Baseline syndesmophytes/ankylosis | 1.19 (1.11 – 1.27) | <0.0001 |
| Model D | ||
| Baseline type B CIL | 2.17 (0.64 – 7.40) | 0.21 |
| Baseline fat lesion | 3.98 (1.74 – 9.08) | 0.001 |
| Baseline syndesmophytes/ankylosis | 1.19 (1.10 – 1.28) | <0.0001 |
CIL, corner inflammatory lesion.