Page 33 - SPONDYLOARTHRITIS IN COLOMBIA -

Imaging and lab characteristics
HLA-B27 testing was performed in 77% of the patients and 43.9% of them were positive. Regarding imaging, radiographs and MRI-SI were performed in 59% and 24% of the patients respectively. Splitting by disease subtype, the subgroup of patients in which most information was available was the subgroup of AS: radiographs: 92%; MRI: 26%; and HLA-B27: 80%. The subgroup in which least data were available was the subgroup of ReA: radiographs: 38%; MRI: 13%; and HLA-B27: 71%. In the subgroup of patients with chronic back pain (past history or current), radiographs were required as complementary testing in the majority of these patients (85%) and less frequently MRI-SI (28%). Patients with back pain in which imaging was performed, sacroiliitis on radiographs was detected in 54%, and MRI-SI inflammation in 56%. HLA-B27 testing was performed in 80.1% of these patients, and 43.6% of them were positive.
Performance of classification criteria
In total, 538 of the 581 patients (93%) in the cohort met at least one criteria set: ASAS: 81% (43.4% had axSpA, and 37.2% had pSpA); ESSG: 74%; Amor: 53%; and 69% (n=399) of patients met two criteria sets. When MRI was used as a substitute for radiographic sacroiliitis, the classification rate slightly increased to 75% for ESSG and to 58% for Amor. The performance of each criteria set is summarized in table-2. All patients with a diagnosis of AS by the rheumatologist fulfilled at least one of these three criteria sets, and 56% of these patients fulfilled all three criteria sets. In particular, patients with a clinical diagnosis of ReA (40%) and uSpA (35%) often fulfilled all three criteria sets, which may reflect the degree of overlap in the criteria for those disease subtypes. Evaluating the ESSG and Amor-criteria, the addition of MRI instead of radiographs to the criteria set did not substantially modify the level of concordance with the ASAS-criteria, which increased from 71.3% to 72.5% and from 57.1% to 60.9% respectively (table-3).
31