The aim of this study was to compare PET with (18)F-FDG PET, in-line PET/CT, and software fusion of independently acquired CT and PET scans for staging of recurrent colorectal cancer (CRC).
Methods: Fifty-one patients with suspected recurrent CRC were studied with in-line PET/CT. Thirty-four of these patients underwent an additional CT scan of the chest or abdomen within 4 wk of PET/CT. Software fusion of PET and CT was performed using a fully automated, intensity-based algorithm. The accuracy of the coregistration of PET and CT scans was evaluated by measuring the distance between landmarks visible in the PET and CT images. Histologic evaluation and follow-up for 6 mo served as the gold standard for the presence or absence of recurrent CRC.
Results: On a patient basis, the accuracy of staging was significantly higher for in-line PET/CT than for PET (88% vs. 71%, P = 0.01). Software fusion of the independently acquired PET and CT images was unsuccessful in 8 patients (24%). In the remaining patients, the mean distance between 62 landmarks visible in PET and CT was 12.9 +/- 7.9 mm, whereas it was only 7.7 +/- 4.7 mm for in-line PET/CT (P < 0.001).
Conclusion: In patients with suspected recurrent CRC, in-line PET/CT significantly improves staging compared with PET alone. Due to its high failure rate, software fusion of independently acquired PET and CT studies cannot be considered to represent an alternative to in-line PET/CT.