Temporal artery biopsy and temporal artery ultrasound inter-rater agreement for the diagnosis of giant cell arteritis: an ancillary analysis from the multicentre prospective ECHORTON study.

Temporal artery biopsy (TAB) and, more recently, temporal artery ultrasound are recommended for the diagnosis of giant cell arteritis (GCA). The inter-rater agreement for TAB is poorly reported, and agreement for ultrasound is variable. A prospective study, the ECHORTON study, evaluated a diagnostic strategy for GCA that used temporal artery ultrasound as the first-line diagnostic test and TAB for ultrasound-negative patients. Clinical expertise served as the reference method. We propose assessing the inter-rater agreement in interpreting TAB and ultrasound images collected in this study. From 2016 to 2020, 165 patients with high suspicion of GCA were enrolled in the ECHORTON study at four general hospitals and two university hospitals. Pathologists and vascular medicine experts independently reviewed TAB and ultrasounds, respectively. The TAB samples were stained with eosin and silver and classified as positive, negative, or equivocal for GCA. Ultrasound results were considered positive when halos were detected around the lumen of both temporal arteries using 9-4 to 18-6 MHz linear probes. This study involved double-blind analysis of 4384 sections from 77 TAB, and 5781 images from 132 ultrasound scans. Kappa coefficients were 0.75 [95% CI: 0.56-0.94] for temporal artery biopsy and 0.73 [95% CI: 0.56-0.90] for temporal artery ultrasound. The reproducibility of interpretations showed heterogeneity across centres, with agreement ranging from fair to excellent. Overall, both TAB and ultrasound demonstrated good inter-rater agreement for GCA diagnosis, though agreement levels varied from fair to excellent across hospitals. Trial registration: The ECHORTON study was registered in ClinicalTrials.gov under the number NCT02703922 on March 3, 2016.