C. Peddes

IOTA simple rules for discriminating between benign and malignant adnexal masses: prospective external validation

To determine the diagnostic performance of International Ovarian Tumor Analysis (IOTA) ‘simple’ rules for discriminating between benign and malignant adnexal masses.

Assessing the reproducibility of the IOTA simple ultrasound rules for classifying adnexal masses as benign or malignant using stored 3D volumes

OBJECTIVE To analyze the reproducibility of the IOTA simple ultrasound rules for classifying adnexal masses as benign or malignant among examiners with different level of expertise using stored 3D volumes of adnexal masses. STUDY DESIGN Five examiners, with different levels of experience and blinded to each other, evaluated 100 stored 3D volumes from adnexal masses and looked for the presence or absence of malignant or benign features according to the IOTA definitions. Multiplanar view and virtual navigation were used. All examiners had to assess the 3D volume of each adnexal mass and classify it as benign or malignant. To analyze intra-observer agreement each examiner performed the assessment twice with a two-week interval between the first and second assessments. To analyze the inter-observer agreement, the second assessment from each examiner was used. Reproducibility was assessed calculating the weighted Kappa index. RESULTS Intra-observer reproducibility was moderate or good for all observers (Kappa index ranging from 0.59 to 0.74). Inter-observer reproducibility was moderate to good (Kappa index range: 0.46-0.67). CONCLUSIONS The simple rules are reasonably reproducible among observers with different level of expertise when assessed in stored 3D volumes.

Transvaginal sonography in diagnosis of vesicoperitoneal fistula due to deep infiltrating endometriosis as a cause of uroperitoneum

ultrasound examination revealed bilateral equinovarus deformity, short femurs, fetal hydrops and persistent retroflexion of the fetal head (Figure 1). The cervical spine and occipital bones were incompletely visualized secondary to persistent fetal position, and no anterior neck mass was detected, though the mandibular skin was in direct continuity with that of the chest due to the persistent neck positioning. Amniocentesis returned a fetal karyotype of 47,XX,+21, with amniotic fluid alphafetoprotein of 0.69 multiples of the median. Testing for intrauterine infection was negative. Persistent retroflexion of the fetal head and worsened hydrops were noted 1 week later and the patient opted to proceed with termination of pregnancy. At the time of intrathoracic potassium chloride injection the following week, the classic finding of fixed, retroflexed fetal head was confirmed (Figure 2). Following uncomplicated induction of labor with misoprostol, postmortem morphology assessment revealed a normal fetal head and neck, with no evidence of an anterior neck mass, occipital bone defect, cervical dysraphism or fixed retroflexion. Formal skeletal films and neuroautopsy confirmed normal bony spine development. Differential diagnoses for a fixed retroflexed fetal head include anterior neck masses (goiter, lymphangioma, teratoma or anterior hygromas), Klippel–Feil syndrome, Jarcho–Levin syndrome and iniencephaly. The phenotype of Klippel–Feil syndrome is caused by failure of segmentation of cervical vertebrae, resulting in cervical fusion, a shortened neck and limited head mobility1. Jarcho–Levin syndrome is autosomal recessive with findings of wedge-shaped vertebra, neural tube defect, rib anomalies and shortened neck2. Though diagnosis of iniencephaly is assured with the classic triad of occipital bone defect, cervical dysraphism and fixed retroflexion of the fetal head, the former two criteria can be difficult to identify with standard two-dimensional ultrasound examination3,4. Indeed, fixed retroflexion of the fetal head with confluence of the skin covering the mandible and chest remains the single most consistently reported ultrasound finding in cases of iniencephaly4–7. We report a case of prenatal diagnosis of a fixed retroflexed fetal head, with associated confluence of the skin covering the mandible and chest, confirmed in our unit on three separate occasions, which was followed by a normal postmortem morphology assessment, skeletal survey and neuroautopsy. We thus suggest that this prenatal finding in the absence of anterior neck mass should not be considered pathognomonic for postnatal bony fusion.

OC11.01: Prevalence of deep endometriosis in patients of a second level ultrasound laboratory and tissue US characterization of nodules using VOCAL

Objectives: To generate physical fetal models using images obtained by 3-dimensional ultrasound (3DUS), magnetic resonance (MR) and computed tomography (CT) to guide additive manufacturing (AM) technology. Methods: Images from 45 fetuses, including 4 sets of twins, were used. Twenty-three fetuses were normal and evaluated only by 3DUS. Twenty-two cases had complications such as conjoined twins; tumors; aneuploidy; skeletal; central nervous system; facial or thoracic defects. Scans were performed using high-resolution 3DUS. In cases of abnormalities, MR and CT were performed on the same day as 3DUS. The images obtained with 3DUS, MR or CT, were exported to a workstation in Digital Imaging and Communications in Medicine format. A single observer performed slice-by-slice manual segmentation using a digital high definition screen. Software that converts medical images into numerical models was used to construct virtual 3D models, which were physically realised using AM technologies (SLA Viper, Objet Connex 350, ZCorp 510 or FDM Vantage). Results: The main outcomes presented were the possibility to create 3D virtual and physical models from 3DUS, MRI or CT both separately and also in various combinations. AM systems allow the conversion of a 3D virtual model to a physical model in a fast, easy and dimensionally accurate process. They were remarkably similar to the postnatal appearance of the aborted fetus or newborn baby. Conclusions: This study introduced the innovative use of AM models into fetal researches. The results suggest a new possibility for educational purposes or better interaction between parents and their unborn child during pregnancy. Normal fetus (29 weeks): Virtual and physical model built in a powder-based system.

The Ovarian Endometrioma: Clinical Setting and Ultrasound Findings

Ovarian endometrioma is defined as a pseudocyst arising from growth of ectopic endometrial tissue. The typical features of endometriomas are diffuse low-level internal echoes (“ground glass”) in the absence of particular neoplastic features and with a clear demarcation from ovarian parenchyma. Several studies report very high values of specificity with values of sensitivity usually ranging from 87 to 77 %.

OC21.05: Low rate of inconclusive findings in less expert operators using ultrasound simple rules for the diagnosis of ovarian cancer

and LR2, the simple rules and the RMI were applied to each of the three groups. Sensitivity, specificity, positive and negative likelihood ratio (LR+, LR−), Diagnostic Odds Ratio (DOR), and area under the receiver operating characteristic curve (AUC) were used to describe diagnostic performance. The gold standard was the histological diagnosis of the surgically removed adnexal mass. Results: The rate of invasive malignancy was 10% in the small tumors, 19% in the medium sized tumors and 40% in the large tumors; 11% of the large tumors were borderline tumors versus 3% and 4% of the small and medium sized tumors. The type of benign histology also differed between the three subgroups. For all methods, the sensitivity with regard to malignancy was lowest in the small tumors, while the specificity was lowest in the large tumors; the DOR and the AUC were highest in the medium sized tumors. The AUC was largest in tumors with a largest diameter 7–11 cm. Conclusions: Tumor size affects the ability of subjective assessment, LR1 and LR2, the IOTA simple rules and the RMI to correctly discriminate between benign and malignant adnexal masses.

OP29.09: Accuracy and reproducibility of ultrasonography for the diagnosis of ovarian cancer in operators with different degrees of experience

according to the IOTA protocol. A proper counselling on a low ovarian cancer risk was given and an adequate informed consent was obtained. Sonographic follow-up was proposed at 3 and 9 months, than yearly. Surgery was proposed in case of changes of ovarian cyst volume (>50%) and/or sonographic parameters at TVS-CD. Demographic, medical, sonographic and pathologic data were recorded. Results: 98 post-menopausal patients with 106 ovarian cysts were enrolled. Median age (IR) was 69 years (59–74). Median years after menopause (IR) were 18 (8–24). Median BMI (IR) was 26 kg/m2 (24–29). Mean parity (±sd) was 2 ± 1.8 women reported bilateral cysts. 53 cysts were unilocular, 53 multilocular. Median cyst diameter (IR) of unilocular and multilocular ovarian cysts was 50 mm (34–56). At a median follow-up period of 36 months, surgery was performed in 28 cases: 14 patients required immediate removal, 5 drop-out, 9 volume changes. One successfully managed cardiac arrest occurred during laparoscopic procedure. No malignant lesions were reported at histology. Conclusions: Sonographic follow-up might be a useful option in the clinical management of incidental unilocular >5 cm and multilocular <7 cm ovarian cysts in post-menopausal women. Larger and longer multicentres studies with strict sonographic parameters are needed to support this potentially safe conservative management.

OP29.06: Perception error and levels of certainty in diagnosing ovarian cancer in operators with different degree of experience

Objectives: To evaluate, sonographic, power-Doppler features and histological findings in cases of small unilocular-solid cyst in a normal-size ovary. Methods: All consecutive women with small ovarian unilocularsolid cyst in normal-size ovary detected by transvaginal ultrasound, were included in the study. Patients had either follow-up ultrasound examination every three months or underwent surgical removal of the cyst. CA 125 was also evaluated. Histological findings were reviewed. Results: A total of 45 cases were identified. The median age of the patients was 48 years (range 25–77). CA 125 was elevated (>35 UI/mL) in 6/45 (13%) cases. The median size of the cyst was 18 mm (range 11–30). The vascularization was present in 20/45 (44%) cases. 15/45 (33%) women received followup ultrasound examination every three month. 30/45 (67%) undervent laparoscopic removal of the cyst. Hystological diagnosis of malignancy (three carcinoma, and six borderline tumors) was recorded in 9/45 (20%) cases and of benign lesion in 21/45 (47%). Conclusions: These data suggest that the presence of a papillary projection is more important than the size of the cyst. Small unilocular solid cysts even in a normal-size ovary have substantial risk of malignancy.