PubMed1998
OBJECTIVES: The aim of this study was to evaluate and compare the use of B-mode transvaginal, utero-ovarian Doppler and intratumoural colour Doppler imaging in the diagnosis of ovarian tumours in order to establish whether they are organic or functional, and malignant or benign in nature prior to surgery. STUDY DESIGN: A series of 213 women between the ages of 15 and 87 presenting ovarian tumours were examined with pulsed and colour Doppler and the functional state of the ovary at the time of the exploration was also established. Size and morphology were evaluated through a new scoring system called sonographic index. This score is calculated as the sum of the points corresponding to tumoural volume (< or = 8 ml = 1 point; > 8 ml and < or = 100 ml = 2 points; > 100 ml = 3 points) and the points corresponding to morphological characteristics (negative echogenicity = 1 point; mixed echogenicity = 2 points; multilocular morphology = 3 points; complex morphology = 4 points; positive echogenicity = 5 points) varying between 2 and 8 points. The flow velocity waveform (FVW) of the utero-ovarian and intratumoural arteries were obtained and analyzed after assessing their resistance indices. Other parameters studied were the presence and location of the colour signal in the intratumoural arteries. RESULTS: Of the total, 84 of the tumours were found to be functional and had disappeared by the time later examinations took place. Surgery was performed in 129 of the patients, and posterior histopathological studies permitted classifying the tumours as benign in 107 cases and malignant in 22. The sonographic index was significantly higher in the organic (4.9 +/- 1.5) and malignant (6.9 +/- 1.0) tumours than in the functional (3.6 +/- 1.5) and the benign (4.1 +/- 1.2) tumours. The utero-ovarian and intratumoural blood flow RIs in the malignant tumours (0.48 +/- 0.12 and 0.43 +/- 0.08, respectively) was significantly lower than in the benign tumours (0.83 +/- 0.12 and 0.58 +/- 0.15, respectively). There were no significant differences in the utero-ovarian and intratumoural blood flow RIs when organic and functional tumours were compared. In all of the malignant tumours a colour signal was identified; it was centrally located in 90% of the cases. A colour signal was observed in 83% of the functional tumours, and in 52% of the organic benign tumours. In the benign tumours the colour signal was peripherally located in 98% of the cases. The variables for sensitivity, specificity, positive predictive value, negative predictive value and precision when organic tumours were diagnosed were 82.9, 65.5, 78.7, 71.4 and 76.1 for a sonographic index where the score was > or = 4; the values for the utero-ovarian RI were 48.4, 90.4, 88.4, 53.6 and 65.1 (cut-off 0.55 > or = RI > or = 0.90); the values for the intratumoural RI were 56.5, 68.6, 54.2, 70.6 and 63.8 (cut-off < or = 0.55). When malignancy was diagnosed these variables were 95.5, 82.2, 38.2, 99.4 and 83.6 for the sonographic index (score > or = 6); 90.9, 98.9, 90.0, 98.9 and 98.1 for the utero-ovarian RI (cut-off < or = 0.55) and 80.0, 66.7, 33.3, 94.1 and 69.0 for the intratumoural RI (cut-off < or = 0.50). The number of false positives diagnosed on the basis of the intratumoural RI decreased when cases that had been evaluated during the luteal phase were excluded from the study. Notwithstanding, the phase of the cycle does not seem to interfere with the diagnosis when utero-ovarian Doppler imaging is used. CONCLUSIONS: B-mode transvaginal ultrasonography and utero-ovarian Doppler velocimetry seem to complement each other to aid in differentiating between organic and functional ovarian tumours. Transvaginal ultrasonography and Doppler imaging provide good results for the diagnosis of malignancy, although the utero-ovarian RI reduces the number of false positives that occur with ultrasonography. Our group, therefore, believes that the combined use of these techniques is beneficial in the clinical d