24–28 August 2008, Chicago, USA Oral communication abstracts sonohysterography & invited to enter the SISES study. Ninety- eight patients were prospectively recruited. SISES was performed at the end of sonohysterography using a Uterine Explora II Curette at the sight of the endometrial abnormality or at a representative site in the uterine cavity if the endometrial cavity was normal. The histopathologic diagnoses of EMB & SISES were compared to a gold-standard pathological diagnosis obtained on hysteroscopy- D&C or hysterectomy specimen. Kappa values for the two different sampling techniques were compared. Results: Eighty-eight SISES samples were obtained. The Kappa between SISES & final pathology was 0.37 (0.20, 0.53). When comparing EMB to final pathology diagnoses, the Kappa was only 0.13 (−0.03, 0.29). EMB underestimated the incidence of pathology, especially when focal lesions & malignancy were present. This difference was statistically significant, with a P-value of 0.0448. Sonohysterography outperformed both sampling techniques, with a Kappa of 0.74 (0.63, 0.86). Classification n Kappa (95% CI) EMB 51 0.13 (−0.03, 0.29) SISES 88 0.37 (0.20, 0.53) Sonohysterography 91 0.74 (0.63, 0.86) Conclusion: SISES is superior to EMB in diagnosing endometrial pathology in peri- & postmenopausal women with abnormal uterine bleeding. OC189 Saline contrast sonohysterography with liquid-based endometrial cytology in post-menopausal patients: Preliminary results F.P.G. Leone 1 , C. Marciante 1 , F. Pagano 2 , M. C. Antonacci 2 , G. Tosi 1 , G. Vago 2 , E. Ferrazzi 1 1 Dept Obstetrics & Gynecology, DSC L. Sacco, University of Milan, Milan, Italy, 2 Dept Pathology, DSC L. Sacco, University of Milan, Milan, Italy Objectives: To assess the diagnostic accuracy of saline contrast sonohysterography (SCSH) with liquid-based endometrial cytology compared to biopsy in post-menopausal patients. Methods: 32 consecutive post-menopausal patients enrolled for uterine surgery (hysterectomy, hysteroscopy) for benign (uterine prolapse and/or symptomatic endometrial polyp, typical hyperplasia, atrophy) and malignant (endometrial atypical hyperplasia, cancer) lesions were prospectively recruited. All patients were pre- operatively assessed by SCSH performed using the Lollo-cath, a 2.4 mm (8Fr) bioptic intrauterine catheter, and injecting 10ml of saline sterile solution. The retrieved fluid at the end of SCSH, by a syringe vacuum aspiration, was considered for liquid-based endometrial cytological analysis. Pathologic reports at hysteroscopy or hysterectomy were considered the gold standard. Results: Median age (IR) was 65 years (61–74). Median years after menopause (IR) were 16 (11–20). 4 patients were in TMX, 2 in HT. Mean BMI (+−sd) was 29 kg/m 2 +−7. Mean parity (+−sd) was 2+−2. 17 patients reported post-menopausal bleeding, 15 were asymptomatic. Median endometrial thickness (IR) at TVS was 6 mm (3–10). Sonohysterography was performed in all patients but one related to cervical stenosis (endometrial atrophic glandulocystic polyp at histology). In two cases endometrial cytological analysis was inadequate, in all four cases endometrial typical hyperplasia was misdiagnosed. No pre- and malignant lesions were misdiagnosed. Conclusions: Saline contrast sonohysterography with liquid-based endometrial cytology performed by using the Lollo-cath showed to be an accurate diagnostic tool in post-menopausal patients. Particularly, it might select high risk cases for endometrial cancer, and reduce unnecessary invasive and expensive procedures. OC190 Three-dimensional transvaginal sonography in the evaluation of the uterine cavity M. Momtaz , A. Ebrashy, M. M. Aboulghar, A. Z. AlSheikha, A. AlKateb, S. Negm, R. Kamel Fetal Medicine Unit, Cairo University, Cairo, Egypt Objectives: To evaluate the use of three-dimensional transvaginal sonography (3D TVS) in the assessment of uterine cavity pathology and congenital anomalies. Materials and Methods: One hundred and twenty three patients were studied. All cases were subjected to 2D TVS, 3D TVS and hysterosalpingography (HSG). Hysteroscopy/Laparoscopy were done for all patients to whom ultrasound and radiology results were compared. Results: Out of 123 cases studied, 102 were found to have: uterine cavity anomalies (n = 38, Group I), uterine myomas, and polyps (n = 52, Group II) and intrauterine adhesions (n = 12, Group III). In Group I, sensitivity, specificity and predictive values were higher for 3D TVS compared to 2D TVS or HSG. Also, the positive and negative likelihood ratios were strong and comparable for 3D TVS and 2D TVS + HSG. In Group II, 3D TVS showed highest sensitivity, specificity and predictive values which were similar to 2D TVS + HSG but higher than each technique alone. In Group III, all tests showed low sensitivity and positive predictive value. The negative likelihood ratio was poor for all methods. Conclusions: The use of 3D TVS is a valuable non-invasive method for assessment of uterine cavity pathology and is more useful than HSG and 2D TVS combined in cases of uterine anomalies, submucous myomas and polyps. However, its value in the assessment of intrauterine adhesions is limited. OC191 Early diagnosis in ovarian cancer: Role of transvaginal color Doppler ultrasound. A nine years experience M. A. Pascual 1 , L. Hereter 1 , B. Graupera 1 , F. Tresserra 2 , I. Rodriguez 1 , S. Dexeus 1 1 Department of Obstetrics, Gynecology and Reproduct, Institut Universitari Dexeus, Barcelona, Spain; 2 Department of Pathology * , Institut Universitari Dexeus, Barcelona, Spain Background: Ovarian cancer mortality remains high mainly due to late diagnosis. Since 1999 we have performed color doppler transvaginal (CDTV) ultrasound as a form of screening for ovarian cancer. The purpose of this paper is to assess the efficiency of CDTV ultrasound to detect early stages of ovarian cancer. Materials and Methods: Screening CDTV ultrasound has been annually performed in asymptomatic women without a family history of ovarian cancer. When abnormalities were detected, the procedure was repeated after 4–6 weeks. If the findings remitted, the study was repeated after one year. If the abnormality persisted, the study was complemented with tumor markers, CT scan and laparoscopy. Results: A total of 112.190 screening CDTV ultrasound were performed. In 34patients a malignant tumor was diagnosed and histologically confirmed, fifteen of them being borderline lesions. Twenty two of 34 patients (67.6%) had normal levels of CA 125. The mean age of these patients was 45.7 years (±13.5). Twenty-five of these tumors (73.5%) were in stage I, 1 in stage II, 7 in stage IIIc and one was a metastatic melanoma. Conclusions: Although consensus about the benefits of transvaginal ultrasound as a screening procedure is not unanimous, our data suggest that tumors detected in patients screened with transvaginal ultrasound are in earlier stages. Additional randomized studies are needed to support these preliminary findings. Ultrasound in Obstetrics & Gynecology 2008; 32: 243–307 305