Comparison of MRI and 18F-FDG PET/CT in the preoperative evaluation of uterine carcinosarcoma

• Preoperative 18F-FDG PET/CT or MRI is not routinely recommended.
• 18F-FDG PET/CT may not replace lymphadenectomy or MRI for predicting LN metastases.
• 18F-FDG PET/CT can be useful for detecting extrauterine disease.

ObjectiveTo compare the validities of magnetic resonance imaging (MRI) and 18F-fluoro-deoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) in preoperative evaluation of uterine carcinosarcoma.MethodsPathologic results of primary tumor lesions and paraaortic and pelvic lymph node (LN) areas were compared with the preoperative image findings. Differences in the validity parameters of both images were compared using McNemar test.ResultsFor detecting primary tumor lesions (n = 56), the sensitivity, specificity, accuracy, positive predictive value, and negative predictive value for 18F-FDG PET/CT versus MRI were 98.1% versus 98.1% (P = 1.000), 33.3% versus 100% (P = 0.157), 94.6% versus 98.2% (P = 0.500), 96.3% versus 100%, and 50% versus 75%, respectively. For paraaortic LN areas, the values were 77.8% versus 51.9% (P = 0.016), 90.2% versus 100% (P = 0.025), 85.9% versus 83.3% (P = 0.774), 80.8% versus 100%, and 88.5% versus 79.7%, respectively. For pelvic LN areas, the values were 61.1% versus 50% (P = 0.125), 86.8% versus 89.5% (P = 0.727), 78.6% versus 76.8% (P = 0.774), 68.8% versus 69.2%, and 82.5% versus 79.1%, respectively. For extrauterine disease, the patient-based values for 18F-FDG PET/CT were 100%, 78.9%, 85.7%, 69.2%, and 100%, respectively.ConclusionIn patients with uterine carcinosarcoma, 18F-FDG PET/CT is comparable to MRI in detecting primary uterine lesions. For predicting LN metastases, though 18F-FDG PET/CT might be insufficient for replacing lymphadenectomy or MRI, it might allow lymphadenectomy to be omitted in poor surgical candidates. For detecting extrauterine metastases, it could also be useful to identify unsuspected disease.