Ab0594 ultrasound intima media thickness cut-off values for cranial and extracranial arteries in patients with giant cell arteritis

BackgroundUltrasound (US) is a valid imaging tool to detect signs of giant cell arteritis (GCA). Although the halo sign has always been considered the most useful finding for GCA diagnosis, modern high frequency transducers are able to precisely measure the intima-media thickness (IMT) of cranial and extracranial arteries. However, data on optimal cut-off values for IMT to differentiate patients and controls in clinical practice are limited.ObjectivesTo determine the optimal cut-off value for IMT of cranial and extracranial arteries in patients with suspected GCA.MethodsThis is a retrospective observational study of patients referred to our US fast-track clinic with suspected GCA. All patients underwent bilateral US examination of the cranial and extracranial (carotid, subclavian and axillary) arteries within 24 hours per protocol. The exam was performed using an EsaoteMyLab8 with a 12-18 MHz frequency transducer for cranial arteries and an 8-14 frequency transducer for extracranial arteries. The IMT was measured in gray scale mode and the presence of a non-compressible halo sign was checked in all arteries. The gold standard for GCA diagnosis was clinical confirmation by the referring rheumatologist after 6 months follow-up. Mean IMT values of each artery were compared between patients with and without GCA by independent samples T-test. Receiver operating characteristics analysis was performed and the Youden index was used to determine the optimal cut-off value for IMT of each artery.ResultsOf the 157 patients with suspected GCA (67.5% female, mean age 73.7 years) referred to the fast-track clinic, 47 (29.9%) had GCA clinical confirmation after 6 months. 41 (87.2%) patients with GCA had positive US findings (61.7% had cranial involvement, 44.7% extracranial involvement and 19.1% a mixed pattern of cranial and extracranial arteries). The following IMT cut-off values showed the highest diagnostic accuracy: 0.44mm for the common superficial temporal artery; 0.34 mm for the frontal branch; 0.36 mm for the parietal branch; 1.1 mm for the carotid artery: 1 mm for the subclavian and axillary arteries. The area under the ROC curve of the IMT for a clinical diagnosis of GCA was 0.984 (95% CI 0.959 - 1) for common superficial temporal artery, 0.989 (95% CI 0.976 -1) for frontal branch, 0.991 (95% CI 0.980 - 1) for parietal branch, 0.977 (95% CI 0.961 – 0.993) for carotid, 0.99 (95% CI 0.979 - 1) for subclavian and 0.996 (95% CI 0.991 -1) for axillary arteries (Table 1).Table 1.Optimal IMT cut-off values for cranial and extracranial arteriesArterySidePatients without GCAPatients with GCACut-off (mm)AUC (CI 95%)Sensitivity (%)Specificity (%)Common superficialtemporal artery mm, mean (SD)Right0.33 (0.06)0.68 (0.28)0.430.997 (0.988 -1)10097.1Left0.35 (0.11)0.57 (0.21)0.450.966 (0.905 -1)10092.3Both0.34 (0.08)0.63 (0.25)0.440.984 (0.959 -1)94.795.1Frontal branch mm, mean (SD)Right0.26 (0.05)0.4 (0.18)0.340.994 (0.983 -1)10097.1Left0.27 (0.05)0.4 (0.18)0.340.985 (0.962 -1)10096.1Both0.26 (0.05)0.4 (0.18)0.340.989 (0.976 -1)10096.6Parietal branch mm, mean (SD)Right0.27 (0.05)0.43 (0.18)0.360.994 (0.981 -1)10098.9Left0.27 (0.05)0.41 (0.16)0.360.987 (0.967 -1)10097.6Both0.27 (0.05)0.42 (0.17)0.360.991 (0.980 -1)10098.3Carotid mm, mean (SD)Right0.8 (0.17)0.88 (0.29)10.974 (0.949 – 0.999)10092.6Left0.82 (0.15)1 (0.42)1.20.982 (0.961 - 1)90.996.2Both0.81 (0.16)0.96 (0.36)1.10.977 (0.961 – 0.993)9094Subclavian mm, mean (SD)Right0.74 (0.18)0.99 (0.44)10.987 (0.97 - 1)10093.4Left0.67 (0.17)0.9 (0.35)1.10.991 (0.975 - 1)10098.3Both0.7 (0.18)0.94 (0.4)10.99 (0.979 - 1)10096Axillary mm, mean (SD)Right0.69 (0.16)0.99 (0.5)10.992 (0.982 - 1)10096Left0.67 (0.17)0.99(0.49)10.998 (0.995 -1)10098.3Both0.68 (0.17)0.99 (0.49)10.996 (0.991 -1)10097.1ConclusionDifferent IMT cut-off values for each artery are necessary to establish a correct US diagnosis of GCA. These proposed IMT cut-off values may help to improve the diagnostic accuracy of US in clinical practice.Disclosure of InterestsNone declared