Detection of recurrent ovarian cancer at MRI: Comparison with integrated PET/CT

OBJECTIVE: To compare the diagnostic performances of magnetic resonance imaging (MRI) and positron emission tomography/computed tomography (PET/CT) for the detection of recurrent ovarian tumor. METHODS: Thirty-six patients who underwent primary cytoreductive surgery for ovarian carcinoma received both MRI and PET/CT for the evaluation of ovarian tumor recurrence. Recurrent ovarian tumors in abdomen and pelvis were classified based on site as follows: (1) local pelvic recurrence, (2) peritoneal lesion, (3) lymph nodal metastasis, and (4) distant metastasis. Patient-based and lesion-based analyses were retrospectively performed with the aim of detecting tumor recurrence. For the detection of recurrent ovarian tumors, we compared patient-based and lesion-based diagnostic accuracies of these 2 modalities using the McNemar test. RESULTS: Histopathologic, clinical, and radiological follow-up findings revealed recurrent ovarian tumors in 35 sites of 22 patients. These 35 sites consisted of local pelvic recurrence (n = 15), peritoneal lesions (n = 14), lymph nodal metastasis (n = 4), and abdominal wall metastasis (n = 2). In detecting recurrent ovarian tumor, patient-based sensitivity and the accuracy of PET/CT and MRI were 73% and 91% (P < 0.05), and 81% and 89% (P > 0.05), respectively. In addition, overall lesion-based sensitivity of PET/CT and MRI were 66% and 86%, respectively (P < 0.05). In detecting peritoneal lesions, overall lesion-based sensitivity and accuracy of PET/CT and MRI for peritoneal lesions were 43% and 86%, and 75% and 94%, respectively (P < 0.05). CONCLUSIONS: Magnetic resonance imaging is more sensitive than PET/CT for detecting local pelvic recurrence and peritoneal lesions of recurrent ovarian tumors.