Page 149 - SPONDYLOARTHRITIS IN COLOMBIA -

Considering the marked heterogeneity of axial SpA, this agreement may suggest that the core characteristics of SpA are captured by all criteria. Many Colombian rheumatologists establish a diagnosis of SpA based on these typical clinical characteristics. However, a bias with regard to
additional testing of MRI and HLA-B27 was present in this study. These additional tests were mainly performed based on clinical indication only, especially in those patients with the highest probability of having SpA and in those patients with an uncertain diagnosis. The fact that MRI and HLA-B27 was not available for all patients may have hampered the comparison of the various criteria sets,
especially those sets that include one or both tests. Recently, a systematic literature review has confirmed the good performance and the validity of the various ASAS SpA criteria as tested against the rheumatologist's diagnosis 2. However, in contrast to these validation studies we found a higher
prevalence of peripheral involvement in Colombian patients. This finding has been consistently observed in Latin America for peripheral arthritis as well as for enthesitis 3. Moreover, the frequency of HLA-B27 in our cohort was also rather low in contrast to data reported in other regions including Europe 4. Differences in the prevalence of HLA-B27 in Colombian population, which has a wide geographic and ethnic variation, may explain these findings. Despite these
differences in prevalence of peripheral symptoms and HLA-B27 in our Colombian dataset, the performance of the ASAS SpA criteria is very similar to the published data.
While in clinical practice rheumatologists do not perform MRI or HLA-B27 testing in all patients with a suspicion of SpA, they do require these tests in those patients with the highest probability of having SpA and in those patients in which clinical symptoms fall short to diagnose but suspicion remains. Because of cost, waiting time for MRI and feasibility issues, these tests cannot be performed in all patients suspected of having SpA. In chapter 3, using the same data as employed in the analyses of the study in chapter 2, we have investigated the patient’s characteristics
147