Page 58 - SPONDYLOARTHRITIS IN COLOMBIA -

Table 2. Association of demographic and SpA disease characteristics that independently prompt to ordering SI-MRI. Results from a multivariable logistic regression model
Explanatory variables
Male gender (male vs female) Number of symptoms at presentation (per additional symptom)
Disease duration (per year)
Chronic back pain (yes vs. no) Inflammatory back pain (yes vs. no) Arthritis (yes vs. no)
Enthesitis (yes vs. no)
Uveitis (yes vs. no)
Psoriasis (yes vs. no)
Infection history (yes vs. no) Diagnostic subtype
SI-MRI, sacroiliac joints magnetic resonance imaging
SI-MRI ordered OR (95% CI)
0.79 (0.50-1.22) 1.27 (1.10-1.47)
0.90 (0.57-1.41) 1.57 (0.72-3.23) 1.81 (1.13-2.90) 0.85 (0.54-1.33) 1.57 (1.00-2.49) 1.08 (0.48-2.41) 0.25 (0.53-1.20) 0.82 (0.49-1.37) 1.06 (0.78-1.43)
p Value
0.28
≤0.001
0.66 0.26 0.01 0.48 0.04 0.85 0.08 0.46 0.68
Demographic and disease characteristics of patients in whom HLA-B27 was ordered
In univariate analysis, the number of symptoms at presentation, the presence of IBP, uveitis, lower CRP, ESR and ASDAS-CRP were associated with a request for HLA-B27 testing.
In multivariable analysis, the single factor remaining independently associated with ordering HLA- B27 was the number of symptoms at presentation (OR = 1.45 per additional symptom, CI 95% [1.24– 1.71], p < 0.001). Although uveitis was associated in the univariable analysis with a request for HLA- B27, it just missed statistical significance in multivariable analysis (OR = 3.19, CI 95% [0.94–10.87], p =0.06). Results are presented in Tables 3-4.
56