Spondyloarthritis
Figure 1 Joint distribution, AUC values, threshold, sensitivity and specificity of the evaluated scores. SONAR-4 score includes DIP2, DIP3, MTP1 and the knee joints, all of which showed significant p values in the regression analysis. AUC, area under the curve; DIP, distal interphalangeal; MTP, metatarsophalangeal; PsA, psoriatic arthritis; PsA-Son22, psoriatic arthritis sonographic 22-joint; SONAR, SONography in Arthritis and Rheumatism.
in the inflammatory US findings, consistent with other studies. 28 The effect of advanced therapies improving US scores of inflammation has also been demonstrated in patients with RA and PsA in a study by Ohrndorf et al , where the PD scores significantly improved after 6 months of therapy. 24 Bone erosions were found to be the least frequent lesion in our patients, consistent with other research, which suggests that erosions in PsA tend to be less frequent compared with RA. 2 29 In addition, NBF was detected more frequently at the DIP and PIP joints, similar to other studies. 30 Within SpA subtypes, PsA tended to show significantly more structural changes (both NBF and erosions) compared with axSpA. This is not surprising, as PsA is characterised by NBF and erosions, as previously described by Aydin et al. 31 However, age may also be a contributing factor, as the PsA cohort was significantly older, and age was found to be significantly associated with NBF in the multivariable regression analysis. Regarding the distribution of synovitis across the different joints, we evaluated 11 joints bilaterally (total of 22) with US based on the PsA-Son22 score, which has been used in several previous studies. We found that the wrist, MCP2, MCP3, DIP2 (hand), DIP3 (hand), MTP1 and knee joints all showed significant discriminatory
capacity or a trend towards significance in univariable analyses. All of these joints have also been found to be the most affected joints in PsA in previous studies, including US evaluation. 3 21 28 Additionally, prior research demon- strated that the MTP (predominantly 1 and 2), wrist, knee and MCP (predominantly 1 and 2) joints contrib- uted most to baseline synovitis severity and were the most responsive to treatment over time. 32 The knee joint showed significant values in univariable (p=0.019) and multivariable analyses (p=0.02), as well as in the LASSO model with an OR of 3.3 for a diagnosis of SpA. A highly significant response to treatment has previ- ously been observed in GS US scores for the knee within 6 months of follow-up. 12 Tang et al also reported that the two most commonly affected joints with US in PsA were the knee and the first MTP joints. 33 While the hand DIP2 joint demonstrated discrim- inative ability in the univariable analysis and LASSO model, the hand DIP3 joint only demonstrated signifi- cance in the univariable analysis. We acknowledge that DIP joints may be inflamed either due to PsA or due to secondary/concomitant OA, but it is currently impos- sible to identify the true driver of inflammation using US imaging alone. 34 This uncertainty also applies to RA-specific US scores, where it is difficult to determine
9
Elsehrawy GG, et al . RMD Open 2026; 12 :e006802. doi:10.1136/rmdopen-2026-006802
Powered by FlippingBook