[18F]FDG positron emission tomography and ultrasound in the diagnosis of giant cell arteritis: congruent or complementary imaging methods?

Autor: Christof Rottenburger, Thomas Daikeler, Elke Schegk, Daniel Staub, Stephan Imfeld, Markus Aschwanden, Christoph Berger
Rok vydání: 2019
Předmět:
Zdroj: Rheumatology (Oxford, England). 59(4)
ISSN: 1462-0332
Popis: Objectives [18F]Fluorodeoxyglucose (FDG)-PET/CT and US are both well established for diagnosing GCA. The present study investigates their accuracy and whether they provide overlapping or complementary information in a cohort of patients presenting with suspicion of GCA. Methods We selected consecutive patients from our cohort of suspected GCA cases that underwent both extended vascular US and PET/CT for diagnostic work-up between December 2006 and August 2012. Results A total of 102 patients were included. Diagnosis of GCA was confirmed in 68 patients and excluded in 34 patients (controls). Vasculitic changes in US were most often found in the temporal artery with 32 positive findings on each side, followed by the popliteal artery (10 right, 9 left) and the subclavian/axillary artery (7 right, 8 left). By contrast, PET/CT showed vasculitis most frequently in the vertebral (23 right, 33 left) and common carotid arteries (32 right, 24 left), followed by the subclavian arteries (16 right, 18 left), and the thoracic (17) and abdominal aorta (23). In 37/68 GCA patients PET/CT and US both revealed vasculitic findings, 11/68 had positive findings in US only and 14/68 in PET/CT only. Specificity of US was higher (one false-positive vs five false-positive in PET/CT). On a single segment level, only 20 of 136 positive segments were positive in both imaging modalities. Conclusion PET/CT measuring vessel wall metabolism and US vessel wall morphology showed a comparable diagnostic accuracy for GCA. However PET/CT and US were often discrepant within single vascular regions. Thus PET/CT and US should be considered as complementary methods, with a second imaging modality increasing the diagnostic yield by 16–20%.
Databáze: OpenAIRE