Purpose
Little is known about the diagnostic concordance of images provided by ultrasound probes with emitting frequencies below or above 20 MHz for the diagnosis of giant cell arteritis (GCA).
Methods
We compared, using Cohen's kappa statistic, data obtained with an 18‐MHz and a 22‐MHz probe for the ultrasonographic evaluation of temporal arteries in 80 patients referred for suspected GCA.
Results
The halo sign was found in 25% of cases with the 18‐MHz probe and in 35% with the 22‐MHz probe. The compression sign was positive in 42% of cases with the 18‐MHz probe and 48% with the 22‐MHz probe. GCA was finally diagnosed in 20 patients (25%). The kappa coefficient of agreement was 0.76 (P < .001) for the halo sign, and 0.75 (P < .001) for the compression sign.
Conclusions
Images obtained by 18 MHz and 22‐MHz frequency probes showed a good level of agreement for the diagnosis of GCA, but the 22‐MHz probe yielded a correct diagnosis of GCA in 3 of the 7 patients in whom examination with the 18‐MHz probe was negative.