To compare the accuracy of whole-body positron emission tomography (PET)using [18F]fluorodeoxyglucose (FDG) with conventional radiological imaging techniques in identifying operable colorectal cancer metastases to the liver.
A double-blind comparative study of FDG-PET as the criterion standard vs conventional radiological imaging methods as the criterion standard, in staging of recurrent colorectal cancer.
Institutional practice in a tertiary referral center.
Thirty-four consecutive patients with suspected colorectal cancer metastases, recruited for the study between May 1993 and October 1994.
Conventional radiological methods of cancer staging included abdominal computed tomography (CT) (n=34), chest x-rays (n=15), and chest CT (n=19) to evaluate extrahepatic disease. Twenty-seven patients were subsequently considered to have apparently isolated cancer metastases to the liver. Anatomical resectability was assessed by magnetic resonance imaging (n=24) or CT angiography (n=3) in all study patients. The FDG-PET studies (n=34) were performed within 8 weeks of conventional radiological imaging.
Main Outcome Measures:
Malignancy of suspected lesions detected by means of FDG-PET and conventional radiological imaging was confirmed by histopathologic examination of resected specimens and percutaneous biopsy specimens and by serial CT scans demonstrating progression of disease.
Unsuspected extrahepatic malignant disease that was missed by conventional radiological imaging was detected by FDG-PET in 11 patients (32%). The PET-detected extrahepatic malignant disease included retroperitoneal nodal metastases (n=6), pulmonary metastases (n=3), and locoregional cancer recurrences (n=2). The additional information afforded by PET consequently had an influence on the clinical management in 10 patients (29%).
The FDG-PET method enabled selection of patients with apparently curable colorectal cancer metastases to the liver for hepatic resection.Arch Surg. 1996;131:703-707