Objective: The aim of this prospective study was to evaluate (18)F-FDG-PET, in comparison with CT, for the detection of peritoneal and retroperitoneal metastases of ovarian cancer.
Methodology: 13 patients with primary (n = 7) or recurrent (n = 6) ovarian cancer underwent an attenuation-corrected (18)F-FDG-PET of the abdomen as well as a contrast-enhanced abdominal CT, followed by surgical staging. For data analysis, the abdomen was artificially divided into six regions (right and left subphrenic region, right and left paracolic gutter, retroperitoneum and central abdomen). All images were reviewed and each region was visually scored on (18)F-FDG-PET as well as on CT. (18)F-FDG-PET results were compared with those of CT, using the surgical data as gold standard.
Results: 73 regions were evaluable surgically and or histologically. Sensitivity was slightly better for CT than for (18)F-FDG-PET (74 vs. 66%). Metastases of <5 mm were missed with both techniques. Specificity, however, was clearly better for (18)F-FDG-PET than for CT (94 vs. 77%), especially in patients with recurrent disease, where postoperative changes (hematomas, adhesions, etc.) caused more false positive results on CT. Retroperitoneal lymph node involvement was found in 3/13 patients. The result of (18)F-FDG-PET for the retroperitoneal lymph nodes was correct in all cases, whereas CT was false positive in 2 patients. (18)F-FDG-PET is relatively inaccurate for the right and the left subphrenic region (missing tumor involvement in 5 patients compared to 2 patients for CT).
Conclusion: Given the low sensitivity of both (18)F-FDG-PET and CT for the detection of peritoneal metastases, surgical staging remains the gold standard. Because of the better specificity, (18)F-FDG-PET might be preferred for evaluating residual or recurrent disease after surgery. (18)F-FDG-PET was more sensitive in the retroperitoneal region than intraperitoneal.
Copyright 2003 S. Karger AG, Basel