Determination of reliability of imaging methods (USG, CT and MRI) in diagnosing and staging of ovarian tumorous in correlation with histopathological analysis

The authors in this prospective study deal with the determination of diagnostic accuracy of imaging methods (conventional and Doppler USG, CT and MR) in diagnostics and staging of malignant ovarian tumours. One hundred and ninety-six women with ovarian and adnexal masses were enrolled in the study who underwent diagnostic imaging modalities prior to planned surgery and subsequent histology. Ultrasonography was performed with the aim to confirm the presence of an ovarian mass and to determine its biological nature, while CT and MR were carried out to specify the morphological and structural characteristics of the primary ovarian tumour and its topo-anatomical relationships to its surroundings, as well as to assess the local or distant secondary dissemination both intrapelvically and intraabdominally. In the detection of primary ovarian tumours all three modalities showed a high diagnostic accuracy under the ROC curve: 0.86 for USG, 0.91 for CT, and 0.90 for MR. For intrapelvic diagnostics of a malignancy, extra ovarian and intraabdominal MR had a higher diagnostic accuracy than USG and CT, however, these differences were statistically significant neither intrapelvically (0.86-0.90) nor intraabdominally (0.91 -0.93). In a paired comparison of diagnostic reliability of the methods there was a statistically significant difference in specificity between USG and MR in favour of USG (P = 0.018), in sensitivity between CT and USG in favour of CT (P = 0.014), and in sensitivity between MR and USG in favour of MR (P < 0.002)