18FDG-PET in 733 consecutive patients with or without side-by-side CT evaluation: analysis of 921 lesions

Side-by-side analysis of CT and conventional (18)FDG-PET in oncological imaging is well established. AIM of this study was to find out which patients or diagnostic groups may benefit the most from the newly introduced integrated PET/CT scanners.

407 consecutively admitted oncological patients with accompanying CT (groups A-D) and 326 patients without CT (groups E-G) were examined by conventional ring PET. Two nuclear medicine physicians and two radiologists assessed each patient's PET and CT scans for pathological lesions with regard to localisation and infiltration of adjacent anatomical structures. Patients without pathological PET findings were assigned to groups A (with CT) or E (without CT). If the localisation and/or extent of a pathological PET focus could only be assessed by taking into account the CT scan, the patient was assigned to group C (with CT) or G (without CT). If PET alone was sufficient for both questions the patient was assigned to groups B (with CT) or F (without CT). If neither method allowed for a precise lesion characterisation, the patient was assigned to group D.

38.6% (A, E) of all patients were PET-negative. PET alone sufficed in 20.6% (B, F). Side-by-side reading of PET and CT was needed for 43.5% (C) of patients referred to PET with a current CT. Side-by-side reading of CT and PET did not suffice for 7.3% (D) of patients in that cohort. A total of 28.2% (G) of the cases without CT would have profited from it. The most frequent oncological diagnoses in group D (PET and conventional CT not sufficient) were bronchial carcinoma with abdominal lesions, while in group G (without CT but CT required) head/neck cancer with thoracic lesions was predominant.

Side-by-side reading of PET and already existing conventional CT failed to yield conclusive data with regard to lesion characterisation in only 7.4% of patients so that PET/CT might have been helpful in these cases. 28.2% of the patients without current CT would have profited from an initial PET/CT examination. On the other hand, 59.2% of all patients (negative PET or PET alone sufficing) did not require a CT for lesion characterisation.