RMD Open
Table 1 Clinical examination of the joints and entheses in patients with PsA and axSpA and HC
P values Overall (across three groups)
HC versus SpA
axSpA versus PsA
Total population SpA
axSpA PsA
HC
Number TJC-22 SJC-22 TJC-28 SJC-28 TJC-68 SJC-66
121
80
39
41
41
<0.001 <0.001 0.040 <0.001 0.001 0.004
1.56 (3.1) 0.54 (1.3) 1.95 (4.4) 0.56 (1.3) 4.27 (7.9) 0.73 (1.7)
2.31 (3.53) 0.80 (1.51)
1.49 (2.8) 0.3 (0.89)
3.10 (3.92) 1.27 (1.82) 3.44 (5.64) 1.27 (1.87) 8.3 (10.46) 1.73 (2.33) 26 (63.4) 22 (53.7) 2.7 (3.43) 1.9 (2.86) 4.2 (5.57)
0.10 (0.3) 0.02 (0.2) 0.17 (0.4) 0.02 (0.2) 0.27 (0.5) 0.02 (0.2)
0.002 0.001
2.86±5.14 2.26 (4.5)
0.306
<0.001 0.001 0.011 <0.001 <0.001 0.045 <0.001 0.001 0.002
0.84 (1.58) 6.32 (9.15) 1.09 (1.93) 49 (61.3) 29 (36.2) 2.8 (3.56) 2.3 (3.19) 4.66 (6.0)
0.38 (1.0) 4.23 (7.1) 0.41 (1.0) 23 (59.0) 7 (17.9) 2.9 (3.7) 2.7 (3.5) 5.1 (6.5)
<0.001 <0.001
Tender joints ≥1*, n (%) 58 (47.9) Swollen joints ≥1†, n (%) 30 (24.8)
9 (22.0)
0.859
<0.001 <0.001 0.002
1 (2.4)
0.001 <0.001 <0.001 <0.001 <0.001 <0.001
SPARCC MASES
2.0 (3.16) 1.6 (2.79)
0.51 (1.2) 0.29 (0.7) 0.73 (1.5)
0.718 0.256 0.504
Total enthesitis score‡ 3.33 (5.3)
All data are reported as a mean (SD) unless otherwise specified. Values in bold are statistically significant. *Tender joints: percentage of participants who had at least one TJC. †Swollen joints: percentage of participants who had at least one SJC (≥1). ‡Total enthesitis score=SPARCC score+MASES score.
axSpA, axial spondyloarthritis; HC, healthy controls; MASES, Maastricht Ankylosing Spondylitis Enthesitis Score; PsA, psoriatic arthritis; SJC, swollen joint count; SpA, spondyloarthritis; SPARCC, Spondyloarthritis Research Consortium of Canada; TJC, tender joint count.
difference observed between axSpA and PsA (p=0.348). NBF was more prevalent in PsA (13.9%) compared with both axSpA (3.7%) and HC (5.0%) (p<0.001). Similarly, erosions were seen more frequently in PsA (2.6%) than in axSpA (0.2%) or HC (0.1%) (p<0.001), as detailed in table 2. ‘Active synovitis’ (GS ≥ 2 and PD ≥ 1) was signif- icantly higher in patients with PsA compared with HC (p=0.02). Regarding total scores, the GS and PD scores were significantly higher in patients with SpA compared with HC, while NBF and erosion scores were higher in PsA compared with other groups (online supplemental table 6). Identification of key lesions and sites distinguishing SpA from controls GS synovitis (p=0.002) and PD (p=0.04) were signifi- cantly associated with a diagnosis of SpA in the univar- iable regression. However, only GS synovitis (p=0.01) retained this significant association in the multivariable regression, adjusting for age and BMI (table 3). Regarding the joints involved (table 4), the combined (summed arithmetically at the joint level) GS and PD synovitis (inflammatory score) of five joints (wrist, knee, hand DIP2–3 and MTP1) showed significantly higher scores in SpA versus HC. Moreover, the inflammatory scores of four joints (hand DIP2–3, knee and MTP1) were significantly associated with the diagnosis of SpA in the univariable regression analysis. Three additional joints (wrist, MCP2 and MCP3) showed a trend towards
significance (p=0.052, 0.077, 0.073, respectively). Inter- estingly, the only joint that retained the significant association for a diagnosis of SpA in the multivariable regression was the knee joint (p=0.02). Knee and hand DIP2 joints were identified as significant predictors for a diagnosis of SpA in the LASSO model (table 5). Diagnostic performance of the ultrasound synovitis scores Based on these findings, we developed and evaluated a new synovitis score that incorporates two lesions (GS synovitis and PD). The proposed score (SONAR-7) includes seven joints (wrist, MCP2, MCP3, hand DIP2, hand DIP3, knee and MTP1) identified as significant or marginally significant in the different regression models (tables 4 and 5). Another, more concise score (SONAR- 4), including only the four joints that showed significant results (table 5) in the regression analysis (hand DIP2, hand DIP3, knee and MTP1), was also evaluated. By assessing the diagnostic performance of these suggested new scores compared with other estab- lished scores (‘PsA-Son22’ and ‘28-joint count’ scores), all scores demonstrated good to excellent discrimi- native capacity between patients with SpA and HC, with area under the curve (AUC) values ranging from 0.785 to 0.831, as illustrated in figure 1. SONAR-7 achieved the highest AUC (0.831), closely followed by the PsA-Son22 score (0.830) and the SONAR-4 joint score (0.828). While the 28-joint score demonstrated a slightly lower discriminative capacity (AUC 0.785, 95%
4
Elsehrawy GG, et al . RMD Open 2026; 12 :e006802. doi:10.1136/rmdopen-2026-006802
Powered by FlippingBook