Positron emission tomography (PET)/computed tomography (CT) is a new imaging modality that provides exact coregistration of anatomic and metabolic data. We have investigated to what degree this new technique might affect the interpretation of PET images in a nonselected group of consecutive cancer patients, reflecting routine condition in a busy cancer center.
Whole-body 2-deoxy-2-[F-18]fluoro-d-glucose (FDG)–PET and PET/CT fusion image sets were compared in 100 consecutive, nonselected patients: 21 with head and neck cancer, 39 with chest malignancies, and 40 with malignancies of the abdomen and pelvis. All studies were performed for primary staging or evaluation of therapy and were interpreted by two nuclear medicine physicians also trained in radiology. Areas of abnormal FDG uptake were identified on PET and graded as likely benign, equivocal, or likely malignant. Positron emission tomography/computed tomography fusion images were then made available, and the initial findings were amended if necessary.
One hundred sixty-six areas with abnormal FDG uptake were identified. Based on PET alone, 51 sites were considered equivocal for malignancy. With PET/CT, the number of equivocal lesions decreased to 24. This difference is more marked in the head and neck as well as the abdomen and pelvis. When the equivocal sites were included in the analysis and grouped with the malignant sites, positive predictive value (PPV) of PET/CT was 89% compared with 75% for PET (p = 0.04).
Combined PET/CT results in increased reader confidence and 53% fewer equivocal readings, as well as improved PPV compared with PET alone.