R. Benzie

Ultrasound diagnosis of fetal macrosomia: a comparison of weight prediction models using computer simulation

To assess the frequency of the diagnosis of macrosomia in relation to differing weight estimation formulae in unselected pregnancies.

OP06.02: The K-P algorithm: new 1st trimester growth assessment technique. Correlation with earliest recorded embryo length, comparison with Robinson's formula

persistence of vaginal bleeding after evacuation. We demonstrated that ultrasound is essential not only in diagnostic of HM but also in follow up and in definition of the high-risk population. Clinical use of such risk assessment permits individualization of follow-up according to the risk of developing subsequent GTN. Patients at high risk for GTN may undergo a closer monitoring whereas those at low risk may benefit of a smaller amount of visits. The extent to which these findings may allow an anticipated search of new pregnancy in low risk cases, remains to be determined.

OP19.04: Intra‐observer agreement in the prediction of pouch of Douglas obliteration using 2‐D off‐line analysis of the TVS ‘sliding sign’

rectovaginal septum. When no lesion was seen, observers were asked to judge if the acquisition of the volume was defective, or if no lesion on the rectovaginal septum was evident; defective acquisition cases were discarded, a total number of 83 cases were evaluated. In order to calculate the performance of the introital 3D US, seven discordant cases were reviewed by a third observer. Interobserver agreement was assessed by calculating kappa index (κ), and Sensitivity, Specificity, PPV and NPV by the three observers were also determined. Results: Interobserver agreement was 0.816 (95% CI [0.69–0.93]) (representing a very good agreement). Sensitivity was 74.1%, Specificity 85.5%, PPV 71.4% and NPV 87%. Conclusions: Our results show that introital 3D US for diagnosis of deep endometriosis of the rectovaginal septum is reproducible with very good interobserver agreement.

OP34.07: The effect of preliminary training on quantitative evaluation of sonographer performance in the fetal morphology ultrasound examination

frequent in the groups with malformations and GDM, but higher as in idiopathic cases. Low birth weight combined with severe polyhydramnios was frequently associated with malformations. Moreover, 21.9% of polyhydramnios with GDM were combined with fetal malformations. Conclusions: Diagnosis of polyhydramnios should prompt glucose tolerance testing, detailed sonography, eventually genetic testing and viral serology. Especially small fetuses with polyhydramnios or GDM should be carefully evaluated for malformations.

OP19.06: The prediction of pouch of Douglas obliteration using off-line analysis of the TVS ‘sliding sign’: diagnostic accuracy and inter-observer agreement

Objectives: To describe ultrasound findings in patients with acute salpingitis and to determine if it is possible using ultrasound to discriminate between acute salpingitis and other painful conditions mimicking clinical symptoms/ findings of acute salpingitis. Methods: 52 patients underwent a standardized transvaginal ultrasound scan before diagnostic laparoscopy because of clinical suspicion of acute salpingitis. The laparoscopist was blinded to scan results. Final diagnosis was based on laparoscopy, histology of the endometrium or other histology where relevant. Results: 29 patients had a final diagnosis of cervicitis (n = 3), endometritis (n = 9), or salpingitis (n = 17), 23 (44%) had a diagnosis unrelated to genital infection. In 4 cases the salpingitis was mild, in 8 moderate, in 5 severe (pyosalpinx). Bilateral adnexal masses and bilateral masses lying adjacent to the ovary were seen more often at scan in patients with salpingitis than with other diagnoses (14/17 vs. 6/35, P = 0.000; 11/17 vs. 6/35, P = 0.001). In salpingitis, the masses lying adjacent to the ovaries were on average 2–3 cm in diameter, solid (n = 14), unilocular (n = 4) or multilocular (n = 3) cystic, or multilocular solid (n = 1), and well vascularized at color Doppler. Spectral Doppler results overlapped between patients with different diagnoses. The sensitivity with regard to acute salpingitis of subjective assessment of scan findings by the sonologist was 82%, specificity 77%, positive and negative likelihood ratio (LR+) 3.6 and 0.23. Those of scan findings of bilateral masses lying adjacent to the ovary were 65%, 83%, LR+ 3.8 and LR− 0.42. The corresponding figures for bilateral adnexal masses were 82%, 83%, 4.8 and 0.22. Conclusions: In patients with clinical suspicion of acute salpingitis, absence of bilateral adnexal masses at scan decreases the odds of acute salpingitis 5-fold.

P04.09: Retained fetal bone following surgical termination at 12–16 weeks gestation: a series of 4 case reports

CT chest abdomen pelvis revealed right pleural effusion, bilateral multicystic ovaries and large ascites. 2.7 litres peritoneal fluid was drained with no malignant cells on cytology. She responded well to supportive measures. Histopathology of product of conception demonstrated oedematous appearance of chorionic villi, normal trophoblast proliferation and no evidence of molar changes. Recent identification of FSH receptor gene mutations which increased FSH receptors sensitivity to structurally similar HCG is responsible for spontaneous OHSS. It induces follicular growth, activates LH receptors on granulosa cells and causes the release of vasoactive molecules. Clinical presentation, biochemical tests and ultrasound findings strongly supported the diagnosis of molar pregnancy but it was a non-molar miscarriage. This case would create awareness of this unusual presentation of spontaneous OHSS mimicking a partial molar pregnancy as it could cause major emotional impact on the patient and financial impact on the healthcare system.

P02.02: Do intrauterine pregnancies of uncertain viability becoming viable after the 1st TVS behave differently to initially viable intrauterine pregnancies?

fetuses with umbilical artery pulsatility (PI) > 95th centile after 36 weeks were analyzed using astria data base records in the fetal medicine unit from 01/01/2008 to 01/04/2010. Results: 40 patients fulfill the criteria for the analysis. All 40 patients delivered live born fetuses. 14 babies were admitted to the neonatal intensive care unit. No neonatal deaths recorded at the discharge of the babies. Conclusions: Term singleton growth restricted pregnancies with umbilical artery Doppler PI > 95th centile, induction of labour using standard prostraglandings regimes seem to be safe regards to neonatal mortality. Further randomised controlled studies need for stasitical evaluation of this hypothesis.

OC05.08: Outcome of the 1st trimester of intrauterine pregnancies of uncertain viability (IPUVIs)

Objectives: To determine the first trimester outcome in women classified with an IPUVI at the primary scan. Methods: Prospective observational study. All pregnant women presenting to the EPU, between Nov 2006 and Jan 2010, underwent a transvaginal scan (TVS). Data were collected from all women classified with an IPUVI at 1st TVS. An IPUVI was defined if there was an empty gestational sac of <20 mm diameter; a gestational sac with a yolk sac, but no CRL; or a gestational sac, a yolk sac, and a CRL of <6 mm with no detectable fetal heart rate (FHR). The women were followed up until the outcome was established: viable or non-viable pregnancy at the end of the 1st trimester. Results: 2048 consecutive pregnant women underwent TVS. 268/2048 (13.1%) were classified with an IPUVI. 220 women (82.1%) had no CRL seen at the first scan, 48 (17.9%) had a CRL of <6 mm. For 237 IPUVIs the 1st trimester outcome was known. 124 (52.3%) were viable and 113 (47.7%) had a miscarriage. Of those IPUVIs which had no CRL on the first scan 57% had a viable pregnancy and 43% had a miscarriage. Of the IPUVIs which had a CRL of <6 mm with no FHR at the first TVS 32.6% were viable and 67.4% were non viable at the end of the first trimester. Conclusions: In our study population the rate of miscarriage in the IPUVI group was high, especially in the subgroup that had a small CRL with no FHR demonstrated at the first scan. We are currently developing a model to predict the outcome of the first trimester for IPUVIs.

OP02.02: The K‐P algorithm: a new 1st trimester growth model, comparison with other formulae and correlations with recorded embryonic lengths

Objectives: The aim was to develop a new model to predict 1st trimester outcomes after a single visit in women classified with an intrauterine pregnancy of uncertain viability (IPUVI) at the primary transvaginal scan (TVS). Methods: Prospective observational study. All pregnant women presenting to the EPU, between Nov 2006 and Jan 2010, underwent a TVS. Data was collected from women with an IPUVI at primary TVS. More than 40 historical, clinical and ultrasonographic (US) end points were recorded for analysis. US measurements included gestational sac (GS) and yolk sac (YS) in 3-planes and crown–rump length (CRL). Women were followed up until the outcome was established: viable or non-viable pregnancy at the end of the 1st trimester. Variables for preliminary model development were determined by stepwise logistic regression. Results: 2048 consecutive pregnant women underwent TVS. 268/2048 (13.1%) were classified with an IPUVI. For 237/268 pregnancies (88.4%) the 1st trimester outcome was known (52.3% viable, 47.7% non-viable). 185 data sets have been used for model fitting. The variables used in the model were maternal age, gestational age in days by LMP, CRL in mm, mean GS size in mm, previous normal vaginal delivery and indication for scan: rescan for previous early intrauterine pregnancy. The predictive ability was measured with an AUC of 0.91, sensitivity for viable 81.1%, specificity 85.3% for the preliminary model. Conclusions: We have developed a new model to predict the likelihood of viability at the end of the 1st trimester in women with an IPUVI. We aim to test this model prospectively to evaluate its performance.

OP02.01: New model to predict viability at the end of the 1st trimester for IPUVIs after a single visit to an EPU—preliminary results

Objectives: The aim was to develop a new model to predict 1st trimester outcomes after a single visit in women classified with an intrauterine pregnancy of uncertain viability (IPUVI) at the primary transvaginal scan (TVS). Methods: Prospective observational study. All pregnant women presenting to the EPU, between Nov 2006 and Jan 2010, underwent a TVS. Data was collected from women with an IPUVI at primary TVS. More than 40 historical, clinical and ultrasonographic (US) end points were recorded for analysis. US measurements included gestational sac (GS) and yolk sac (YS) in 3-planes and crown–rump length (CRL). Women were followed up until the outcome was established: viable or non-viable pregnancy at the end of the 1st trimester. Variables for preliminary model development were determined by stepwise logistic regression. Results: 2048 consecutive pregnant women underwent TVS. 268/2048 (13.1%) were classified with an IPUVI. For 237/268 pregnancies (88.4%) the 1st trimester outcome was known (52.3% viable, 47.7% non-viable). 185 data sets have been used for model fitting. The variables used in the model were maternal age, gestational age in days by LMP, CRL in mm, mean GS size in mm, previous normal vaginal delivery and indication for scan: rescan for previous early intrauterine pregnancy. The predictive ability was measured with an AUC of 0.91, sensitivity for viable 81.1%, specificity 85.3% for the preliminary model. Conclusions: We have developed a new model to predict the likelihood of viability at the end of the 1st trimester in women with an IPUVI. We aim to test this model prospectively to evaluate its performance.