Background Reducing the rate of stillbirth is an international priority. At least half of babies stillborn in high-income countries are small for gestational-age (SGA). The Growth Assessment Protocol (GAP), a complex antenatal intervention that aims to increase the rate of antenatal detection of SGA, was evaluated in the DESiGN type 2 hybrid effectiveness-implementation cluster randomised trial (n = 13 clusters). In this paper, we present the trial process evaluation. Methods A mixed-methods process evaluation was conducted. Clinical leads and frontline healthcare professionals were interviewed to inform understanding of context (implementing and standard care sites) and GAP implementation (implementing sites). Thematic analysis of interview text used the context and implementation of complex interventions framework to understand acceptability, feasibility, and the impact of context. A review of implementing cluster clinical guidelines, training and maternity records was conducted to assess fidelity, dose and reach. Results Interviews were conducted with 28 clinical leads and 27 frontline healthcare professionals across 11 sites. Staff at implementing sites generally found GAP to be acceptable but raised issues of feasibility, caused by conflicting demands on resource, and variable beliefs among clinical leaders regarding the intervention value. GAP was implemented with variable fidelity (concordance of local guidelines to GAP was high at two sites, moderate at two and low at one site), all sites achieved the target to train > 75% staff using face-to-face methods, but only one site trained > 75% staff using e-learning methods; a median of 84% (range 78–87%) of women were correctly risk stratified at the five implementing sites. Most sites achieved high scores for reach (median 94%, range 62–98% of women had a customised growth chart), but generally, low scores for dose (median 31%, range 8–53% of low-risk women and median 5%, range 0–17% of high-risk women) were monitored for SGA as recommended. Conclusions Implementation of GAP was generally acceptable to staff but with issues of feasibility that are likely to have contributed to variation in implementation strength. Leadership and resourcing are fundamental to effective implementation of clinical service changes, even when such changes are well aligned to policy mandated service-change priorities. Trial registration Primary registry and trial identifying number: ISRCTN 67698474. Registered 02/11/16. https://doi.org/10.1186/ISRCTN67698474.

be identified and the pericardial effusion resolved by 28 weeks of gestation. The patient had a spontaneous vaginal delivery at 38 weeks of gestation. The newborn weighed 2810gr, had good Apgar scores and uncomplicated postnatal course. At 2 months of age an echocardiogram demonstrated normal ventricles and a small patent foramen ovale. Our case supports a favorable outcome in cases of isolated cardiac diverticulum and large pericardial effusion diagnosed early in pregnancy in the setting of normal NT and normal cardiac function.

Poster abstractsan uneventful spontaneous delivery of a female infant. After birth the neonate had slightly distended abdomen and yellowish watery diarrhoea without meconium passage. An infantogram after birth showed mild bowel gas distension without evidence of intestinal obstruction, small bowel series showed no small bowel transit time delay and diffuse gas distended colon without obstruction. She had hyponatremic metabolic alkalosis and hypochloridemia. The stool chloride concentration was high, 101 mmol/L (reference value < 90 mmol/L), confirming the diagnosis of CCD. She was discharged with sodium chloride for electrolyte correction.Supporting information can be found in the online version of this abstract. P22.07

Part of the placenta was left in situ. Estimated blood loss was 2500mls. Postpartumly the patient had minimal bleeding for about a week and was discharged. The patient returned with increasing brownish vaginal discharge at 5 weeks postnatally. On ultrasound examination, the retained portion of placenta was confirmed to be in the uterine cavity with no vascularity observed between the placenta and the uterine wall on Doppler imaging. Vaginal misoprostol was given followed by manual removal of the retained placental tissue. There was minimal bleeding associated with the procedure. In conclusion, ultrasound is useful in guiding the management of morbidly adherent placenta. When there is no or little vascularity between the placenta and the uterine wall demonstrable postnatally, it may be safe to remove the placenta.

Objective: It has been proposed that thickness of chorioamniotic membranes is a predictor of preterm delivery (UOG 2008; 32 : 205). The objective was to evaluate sonographic membrane thickness (MT) throughout gestation in 3 different sites: chorionic plate (CP), uterine free wall (FW), and cervix. Method: A prospective, cross-sectional study was performed in patients with singleton pregnancies who underwent transabdominal and transvaginal US. MT was measured in 3 sites: CP, uterine FW (area free of placenta), and cervix (at or within 2 cm of internal os). Three measurements were taken per site. Exclusion criteria were: short cervical length (< 2.5 cm), funneling, positive response to transfundal pressure, presence of sludge, vaginal infections, medical complications, and fetal anomalies. Data analysis was done using linear regression. P<0.05 was considered significant. Intra-observer variability was assessed by 1 operator in 10 patients (for each of the 3 sites), and % coefficient of variation (CV) was computed. Result: MT could be measured at the CP and cervix in all patients (n=40). Membranes could not be visualized at the FW in 57.5% (n=23). Mean (± SD) MT (mm) at the CP, FW, and cervix differed significantly among these sites: 1.28 ± 0.17, 0.97 ± 0.14, and 0.55 ± 0.08, respectively (P<0.0001). MT at the CP increased as a function of gestational age (P = 0.005); however, a similar change was not detected for FW or cervix. CV for MT measurements ranged from 9.8 to 14.4% for intra-observer variability. Observations were made between 15-34.4 (median 25.4) weeks. Conclusion: 1) Mean MT throughout gestation differed significantly among CP, FW, and cervix; 2) FW membranes could not be visualized in 57.5%; 3) MT at the CP increased as a function of gestational age; 4) intra-observer variability was acceptable and was similar among sites; 5) reference values reported herein may be used to assess MT in cases of preterm labor and/or suspected intra-amniotic infection/inflammation.