A Simplified PET/CT Measurement Routine with Excellent Diagnostic Accuracy for the Diagnosis of Giant Cell Arteritis

Abstract

We previously proposed standard uptake value (SUV) ratio-based cut-off values for [18F] fluorodeoxyglucose-positron emission tomography/computed tomography (PET/CT) for diagnosing giant cell arteritis (GCA) with high diagnostic accuracy. Here we confirm our findings in an independent cohort and report a simplified procedure for using a SUV ratio to diagnose LV-GCA. Patients with suspected GCA were consecutively included. The ‘peak SUV ratio’ was defined in a two-step approach. First, the vessel with the visually brightest radiotracer uptake in the supra-aortic (SA) and in the aorto-iliofemoral (AIF) region was identified. Here, the maximum SUV of the vessel was measured and divided by the mean SUV of the liver (SUVratio). A ratio >1.0 in the SA or >1.3 in the AIF region was scored as vasculitis. The diagnostic accuracy, sensitivity, and specificity of the ‘peak SUV ratio’ in the SA and AIF region was assessed. From 2015 to 2019, 50 patients (24 female, median age 71 years) with suspicion of GCA were included, 28 patients with GCA and 22 patients with exclusion of GCA. Peak SUV had an AUC of 0.91, a sensitivity of 0.89, and a specificity of 0.73 for diagnosing GCA. Peak SUV accuracy of the AIF arteries was lower (AUC 0.81) than of the SA arteries (AUC 0.95). Our SUV ratio cut-off values for diagnosing GCA are consistently valid, also when applied in a time-efficient clinical procedure focusing on the peak SUV ratio. The diagnostic performance of PET/CT in this validation cohort was even higher, compared to the inception cohort (AUC of 0.83).

Keywords

• giant cell arteritis
• large-vessel-vasculitis
• PET/CT
• imaging
• diagnosis
• SUV