Although
18F-FDG-PET/CT is a useful tool in differentiating benign from malignant lesions, it cannot be used for specifically determining malignant lesions. Some infectious or inflammatory lesions can also cause high
18F-FDG uptake[
20]. For example, Chen et al.[
21] reported a patient with an adnexal mass referred for
18F-FDG-PET/CT scan owing to elevated tumor marker levels. The
18F-FDG-PET/CT scan showed multiple hypermetabolic foci in the peritoneum that mimicked peritoneal carcinomatosis. However, peritoneal tuberculosis was confirmed later. In our study, 3 out of 7 benign lesions were due to pelvic tuberculosis and mimicked peritoneal carcinomatosis. Conversely, most advanced stage ovarian cancers show peritoneal metastasis, and such peritoneal lesions often cause confusion during diagnosis because images obtained for these cancers are similar to those obtained for benign inflammatory conditions involving the gross peritoneal surface, as seen in our series.
Generally, whole-body
18F-FDG-PET/CT examination is performed 1 h after the administration of
18F-FDG. This seems to be ideal taking into account the half-life of the radiotracer, which is 110 min[
17]. However, many different acquisition protocols have been proposed to overcome the aforementioned false-negative or false-positive results obtained with
18F-FDG-PET/CT. Saito et al.[
15] used a protocol involving early-phase and delayed-phase scans (1 h and 2 h after
18F-FDG administration) in 48 consecutive patients with intraductal papillary mucinous neoplasia of the pancreas. In this study, SUV
max increased further in the delayed phase imaging in 92.3% patients with malignant pancreatic neoplasia, and only 60.0% of those with benign pancreatic neoplasia. This result is consistent with that of ours, in which SUV
max increased significantly in delayed phase PET/CT in malignant but not benign lesions. Cheng et al. performed dual-phase
18F-FDG-PET/CT in 35 patients with equivocal infiltrative hepatic lesions on abdominal CT or MRI scans[
16]. In contrast to our study, the authors observed that SUV
max was significantly higher in infiltrative hepatic malignancies than in benign lesions in both early (5.9 ± 5.0 versus 3.9 ± 1.7, respectively) and delayed images (6.8 ± 10.2 versus 4.1 ± 3.9, respectively). Thus, early phase SUV
max of benign and malignant lesions were significantly different. In our study, only the AUC of SUV
max2 was statistically significant in differentiating benign from malignant lesions. This might be attributed to either a different time point (45 min) for the early phase
18F-FDG-PET/CT examination or organ-specific characteristics. Caprio et al. evaluated the performance of dual-phase
18F-FDG-PET/CT in 48 patients suspected with breast lesions[
17]. In their study, dual-time point acquisition of
18F-FDG-PET/CT displayed an accuracy of 85% for lesions with a SUV
max ≥ 2.5, with sensitivity and specificity values of 81% and 100% compared with 63% and 100%, respectively, for the single-time point acquisition. In our study, the sensitivity and specificity of the SUV
max2 were 100% and 57.1%, respectively, with a SUV
max2 > 3.9. The relatively small number of patients and the different time point of the (3 h) in delayed phase
18F-FDG-PET/CT examination might account for the differences in sensitivity and specificity. Yang et al. investigated whether adding delayed phase imaging can improve diagnostic ability of
18F-FDG-PET in evaluating solitary pulmonary nodules[
18]. In their study, combined early and delayed phase scans of 28 patients showed correct diagnosis of the 3 malignant lesions with an initial SUV < 2.5. Despite minor differences in the study design and results compared to those of our study, the abovementioned studies confirmed the clinical usefulness of dual-phase PET/CT in differentiating benign from malignant lesions.
The major limitations of the current study include the small number of patients. Although patients were prospectively enrolled at the beginning of this study, selection bias could have been introduced. In other words, the clinicians’ attitudes to dual-phase PET/CT might affect the referral of patients. Second, this study was performed in a single institution. Differences in image acquisition and interpretation might influence the study results. Therefore, an additional large-scale, prospective study is necessary to check if our findings have an impact on future clinical practice. In addition, cost-effectiveness analysis should also be conducted. Despite these limitations, the findings of the present study provide useful information for the management of patients with equivocal adnexal masses; the sensitivities of both MRI and single-phase
18F-FDG-PET/CT for detecting ovarian cancer have reported to be similar[
22,
23]. Hence, if differential diagnosis of an adnexal mass is indeterminate by MRI, then probably the usefulness of single-phase
18F-FDG-PET/CT are also limited. Dual-phase
18F-FDG-PET/CT may be a better procedure in resolving this ambiguity. Accurate preoperative assessment could help clinicians perform prompt debulking surgery in patients with “true positive” ovarian cancers, and avoid unnecessary surgery in patients with benign conditions.