Background Early diagnosis of axial spondyloarthritis (axSpA) remains a challenge and the potential role of power Doppler (PD) ultrasonography (PDUS) of entheses is still under evaluation.
Objectives To assess: i) the accuracy of PDUS enthesitis at 3 entheseal sites for the diagnosis of axSpA (ASAS criteria) in patients with recent inflammatory back pain (IBP); ii) the prevalence of PDUS enthesitis in each arm of the ASAS criteria; and iii) the prevalence of PDUS enthesitis according to the presence of pain at the same site.
Methods We used patients from the DESIR cohort, with IBP (>3 months and <3 years) suggestive of axSpA according to rheumatologist's opinion1. At baseline, 14/25 centers performed a PDUS examination according to a standardized protocol1, 3 sites were tested bilaterally: Achilles tendon (site A), patellar ligament (site B), common extensor tendon of the elbow (site C). PDUS enthesitis was defined as any vascularised enthesitis by PD within 2 mm from bony cortex. Sample size for the best cut-off for sensitivity (Se) and (Spe) of PDUS for diagnosis of axSpA (ASAS+) was calculated as follow: i) Se>0.8 with a relative error (RE) of 5% and an absolute error (AE) of 5%, requiring 246 patients ASAS+ (274 with 10% of non-interpretable data); ii) Spe>0.7 with an AE of 10% and RE of 5%, requiring 81 patients ASAS- (90 with 10% of non-interpretable data). Thus 364 patients were necessary with a ratio ASAS+/ASAS- of 3/1, for covering both hypotheses.
Results 402 out of 708 included patients were examined by PDUS. After Bonferroni correction for multiple tests, there was no significant difference between patients having received or not the PDUS examination.58 (14%) patients had at least 1 PDUS enthesitis (PDUS+), of which 15 (26%) at site A, 36 (62%) at site B and 20 (34%) at site C. Se of PDUS for the diagnosis of axSpA was 14%, Sp 83%, positive predictive value 69%, and negative predictive value 27%. After Bonferroni correction the lone significant difference between PDUS+ and PDUS- patients was a more frequent history of clinical enthesitis (74.1% vs 47.7%, p<0.001). Table 1 shows the prevalence of PDUS+ patients according to ASAS criteria. Among the 18 patients PDUS+ASAS-, 11/17 (65%) fulfilled Amor's criteria, 16 (89%) ESSG criteria, and 11/17 (65%) both sets of criteria. No statistical relationship was observed between current pain at one site and presence of PDUS enthesitis at the same site.
Conclusions These results raise the potential value of PDUS enthesitis in patients with IBP suggestive of axSpA, but who do not fulfill the ASAS classification criteria for AxSpA.
Dougados M., JBS 2011:78:598-603
Disclosure of Interest None declared