Development and validation of a Pediatric Internationally agreed UltraSound Hip synovitis protocol (PIUS-hip), by the PReS imaging working party.

Abstract

Background: Whilst musculoskeletal ultrasound (MSUS) normal values for examination of the hip joint have been established for healthy children, equivalent values for patients with juvenile idiopathic arthritis (JIA), as well as internationally validated MSUS protocols for the optimal evaluation of synovitis are lacking. This study aimed to develop and validate the most sensitive MSUS protocol for the detection of hip synovitis in JIA.

Methods: In consecutive JIA patients with ≥ 1 clinically affected hip joint, affected and unaffected hips underwent MSUS. Disease, demographic and clinical findings were recorded. Synovitis was graded using the pediatric OMERACT score for B-Mode (BM) and power-Doppler Mode (PD) in the longitudinal and transverse scans and the sensitivity and specificity was analyzed. Additionally anterior recess size (bone to capsula distance), capsula thickness and femoral head cartilage thickness (transverse view) were measured. Published data provided further control data for anterior recess size (children without JIA). Interobserver reliability of BM and PD was tested using Fleiss-Kappa.

Results: 60 patients were enrolled who had 76 hips with and 32 without clinical arthritis. BM was positive (grade ≥ 1) in 74/76 of hips with clinical arthritis (97%, sensitivity 0.97 (0.93-1.0), specificity 0.85 (0.74-0.97) versus 2/32 (6%) in hips without arthritis. PD positivity frequency was 6 (8%) in hips with arthritis versus 0 in hips without. Anterior recess size (mean ± SD) was significantly wider in patients with clinical arthritis (9.9 ± 2.5 vs 5.5 ± 1.3, p-value 0.001). Use of the cut-off of ≥ 7.2 mm resulted in an area under the curve of at least 95%, with a sensitivity of 86% and specificity of 94%. Articular capsula and femoral head cartilage thickness did not differ between patients with and without arthritis. Recess size was comparable in the internal and external control groups (n = 449). Interobserver reliability of BM and PD positivity showed excellent agreement (kappa = 0.85).

Conclusions: The Pediatric internationally agreed UltraSound hip synovitis protocol (PIUS-hip) could be limited to one longitudinal scan including B-Mode scoring plus measurement of anterior recess size for maximal sensitivity and specificity for synovitis.