Diagnostic performance of CT, PET, side-by-side, and fused image interpretations for restaging of non-Hodgkin lymphoma
The purpose of this study was to compare the diagnostic performance of positron emission tomography (PET) alone, computed tomography (CT) alone, side-by-side reading, and fused images for restaging or follow-up of patients with malignant lymphoma.
Fifty patients with histologically confirmed non-Hodgkin lymphoma underwent an 18fluoro-2-deoxyglucose (FDG)-PET scan, followed by a CT scan. CT alone, PET alone, side-by-side reading, and fused images were interpreted separately and visually using a five-point grading scale for the following eight regions: cervical, supraclavicular, axillary, mediastinal, para-aortic to iliac, mesenteric, inguinal, and extra-nodal. Diagnostic accuracy was compared on the basis of the final diagnoses determined by histological confirmation and/or clinical course.
For all regions combined, the interpretation of PET alone (sensitivity = 86.1%, specificity = 99.4%, accuracy = 91.0%), side-by-side reading (96.0%, 99.4%, 98.9%), and fused images (98.0%, 99.4%, 99.2%) yielded significantly higher diagnostic performance than that of CT alone (59.4%, 96.1%, 91.0%; P < 0.001). The cervical, supraclavicular, and extra-nodal regions were more accurately diagnosed with PET (P < 0.05), whereas the para-aortic to iliac regions were diagnosed more accurately with side-by-side reading and fused images than with CT alone or PET alone (P < 0.05).
Although fused images are clinically valuable, side-by-side reading showed equivalent performance, whereas the interpretation of PET alone yielded reasonably high diagnostic performance for restaging or follow-up of patients with malignant lymphoma.