The New ASAS Classification Criteria for Peripheral SpA
The process of developing the new criteria for peripheral SpA (see Figure 4) was similar to that for axial SpA. Two sets of candidate criteria were formulated based on clinical reasoning and then tested in 35 'paper patients', adjusted and validated. Patients without back pain and with peripheral manifestations that usually began before the age of 45 years, but without an established diagnosis, were included. Two hundred and sixty six patients from 24 centers were recruited. Again, in an effort to minimize selection bias patients were enrolled in a strictly consecutive manner, and again clinical diagnosis (SpA or no SpA) by an ASAS rheumatologist was used as the gold standard. A final set of criteria showing the best balance of sensitivity (77.8%) and specificity (82.9%) was decided upon. It consists of peripheral arthritis (usually lower limb predominant and asymmetric) and/or enthesitis and/or dactylitis) plus additional features. These additional features may include one or more of the following: psoriasis, inflammatory bowel disease, preceding infection, HLA-B27, uveitis and sacroiliitis on imaging. Alternatively, they may include two or more of the following: arthritis, enthesitis, dactylitis, history of previous IBP and family history of SpA.
(Enlarge Image)
Figure 4.
Peripheral spondyloarthritis classification criteria.
SpA: Spondyloarthritis.
Data taken with permission from [27].
These new criteria, akin to the criteria for axial SpA, performed better than versions of the Amor and ESSG criteria (which were modified to include MRI findings), particularly in terms of sensitivity. Additionally, a combination of the new criteria for axial and peripheral SpA was compared to the modified versions of the Amor and ESSG criteria in the entire ASAS population of 975 patients. The balance of sensitivity and specificity of the combined new criteria was found to be superior to both of the older criteria sets. These figures for the combined new criteria were sensitivity of 79.5% and specificity of 83.3%, compared with 79.1 and 68.8%, respectively for the modified ESSG criteria, and 67.5 and 86.7%, respectively, for the modified Amor criteria.