ABSTRACTS

Click to increase image sizeClick to decrease image size

Chorionicity in multiple pregnancies: How accurate is ultrasound assessment?

M. Ramalingam, J. Peebles & F. Morrison

NHS Forth Valley, Larbert, UK

Introduction. Multiple pregnancy has increased in frequency as a result of the ageing maternal population and wider availability of assisted reproductive technologies. Higher order pregnancies are associated with increased morbidity and mortality. The introduction of ultrasound (US) has made establishment of multiple pregnancies and their chorionicity more accurate.

Aims

• To determine the accuracy of US in the establishment of chorionicity in comparison to pathological analysis of placenta.

• Sensitivity of US in identifying dichorionic diamniotic (DCDA) and monochorionic diamniotic (MCDA) pregnancies.

• US sensitivity when performed before 14 weeks’ gestation.

Methods. This retrospective study included all patients attending the twins clinic over a 2-year period. A total number of 100 cases were identified. A standardised proforma was used to collect the data.

Results. Among the total number, 78% were DCDA and 21% were MCDA pregnancies. Approximately 15% of the women had a history of assisted conception. The sensitivity of ultrasound for establishing chorionicity was 91%. Sensitivity for DCDA pregnancies were higher compared with MCDA pregnancies (98% vs 69%). A total of 89% of the women had a scan under 14 weeks’ gestation. Surprisingly, sensitivity for US > 14 weeks was 100% and < 14 weeks was only 92%. A small number (17%) of women required a 2nd scan to confirm chorionicity.

Conclusions. This study shows that US assessment of chorionicity has a high sensitivity and specificity. Sensitivity for DCDA pregnancies were higher compared with MCDA pregnancies. The policy of establishing chorionicity by early pregnancy ultrasound assessment is a safe and reliable method.

ßhCGs and the management of early pregnancy complications in Aberdeen

A. Liddiard

Aberdeen Maternity Hospital, Aberdeen, UK

Aims. To audit the use of serum ßhCGs in the management of pregnancy of unknown location in Aberdeen.

Methods. This was a prospective note-based study undertaken over a 2-month period (October and November 2010). Notes of all patients having ßhCGs on the early pregnancy assessment unit and gynaecology wards in that time period were examined, and a standardised proforma was completed. This included: basic patient demographics; presenting symptoms; gestation; risk factors for an ectopic pregnancy; initial ultrasound scan findings; working diagnosis at initial ßhCG and the final diagnosis. A decision was made regarding whether the local guideline on the management of unknown location – based on the RCOG green top guideline for Early Pregnancy Loss (No. 25) – had been adhered to.

Results. A total of 63 patients had one or more ßhCGs in the time period; two sets of notes were unavailable. The mean age was 29 years. A total of 51 patients (84%) presented with bleeding and 45 patients (74%) also had pelvic pain. Six patients (10%) had risk factors for ectopic pregnancy. The commonest working diagnosis at first ßhCG was ‘rule out ectopic’ (59%) and the most frequent final diagnosis was miscarriage (61%). Of the patients, 25% had a pregnancy of unknown location (8% a confirmed ectopic pregnancy) at final diagnosis. The average number of ßhCGs required to make a diagnosis was two (range 0–5) and it was felt that on analysis of the notes, 11 patients (18%) were not managed as per the hospital guideline – with ßhCGs being performed inappropriately or too frequently.

Conclusion. On average, one patient a day commences ßhCG tracking in Aberdeen. Mostly, the tracking is done to rule out an ectopic pregnancy but the commonest diagnosis is miscarriage. A total of 82% of patients are managed according to the local guideline and re-publishing it may improve adherence to it.

Predicting pre-term birth using fetal fibronectin (fFN) in Ninewells Hospital, Dundee

R. Northridge¹, H. Liu², R. Youssef¹ & A. E. Nicoll¹

¹ Ninewells Hospital, ²University of Dundee, Dundee, UK

Aims

• To assess the ability of fFN in predicting pre-term birth (PTB) in Ninewells Hospital, Dundee.

• Specifically, to determine the rate and duration of maternal admission in women that presented with threatened pre-term labour who had fFN.

• To assess whether or not these women received corticosteroids and tocolysis and to assess pregnancy outcomes.

Methods. Data were collected prospectively from all women who presented with threatened pre-term labour between 1 September 2009 and 28 February 2010. Information regarding delivery outcomes was obtained from the maternity case notes and the local maternity database.

Results. A total of 40 women had fFN testing. Six of the 40 (15%) tested fFN-positive. Of the 40 women, 16 (40%) were admitted (fFN-positive = 6/6, 100% vs fFN-negative = 10/34, 29%) (p = 0.002). The mean duration of admission was 2 days (fFN-positive = 2.5 days vs fFN-negative = 1.7 days) (p = NS). A total of 9/40 (22.5%) received corticosteroids (fFN-positive = 5/6, 83% vs fFN-negative = 4/34, 12%) (p = 0.001). No patients received tocolysis.

A total of 8/40 (20%) had spontaneous PTB before 37 weeks’ gestation (fFN positive = 5/6, 83% vs fFN-negative = 3/34, 9%) (p = 0.0005). In our population, fFN has a sensitivity of 62.5% and a specificity of 100% for predicting spontaneous PTB before 37 weeks’ gestation (positive predictive value (PPV) = 83%; likelihood ratio (LR) = 5; Negative predictive value (NPV) = 91%; LR = 0.09).

Conclusions. In our population, the majority of women who present with threatened pre-term labour will deliver at term. A positive fFN test is a moderate predictor of spontaneous PTB before 37 weeks’ gestation. However, a negative fFN test is a strong predictor that spontaneous PTB before 37 weeks’ gestation will not occur. A negative fFN test can reassure women and prevent interventions such as hospital admission and the administration of corticosteroids.

Audit of serum hCG assessment in pregnancy of unknown location (PUL)

M. Cooper

R. Kumar

Ninewells Hospital, Dundee, UK

Background. Pregnancy of unknown location (PUL) is a descriptive term used when nil is seen on a transvaginal ultrasound scan, despite a positive pregnancy test. Its incidence has increased in recent years due to earlier pregnancy detection rates with modern tests. Combining the use of ultrasound scanning with hCG assessment can be used to predict the outcome of PUL. The appropriate use of serum hCG assessment is listed as an auditable standard in the RCOG Guideline No. 25 (Management of Early Pregnancy Loss) and has significant harm and cost implications if used inappropriately.

Aims. The aim was to evaluate the use of serum hCG assessment and management of PUL within the department. The Gold Standard used for comparing our practice was the Association of Early Pregnancy Units (AEPU) algorithm for the management of PUL.

Methods. A retrospective case note review of all patients undergoing serial hCG assessment within the department in a 3-month period (July to September 2010). Management was compared with the Gold Standard, as described above.

Results. In total, 23 patients were included for evaluation. The final diagnoses included five early pregnancies, five ectopic pregnancies and 13 non-continuing (of all origin). Average number of hCG measurements to reach diagnosis was 2.3. The protocol was followed in 15 out of 23 cases (65%). Deviation from protocol was most commonly seen in the group with a suboptimal hCG change, i.e. possible ectopic/persistent PUL group. All five of the early intrauterine pregnancies were managed according to protocol.

Conclusions. The main reasons for deviation included failure of hCG measurement on first visit or failure of repeating serum hCG at 48 hours. Generally, this decision was taken on the basis of clinical judgement or patient choice, e.g. unwanted pregnancy. There is no obvious difference in the management of PUL between the early pregnancy clinic and gynaecology ward out of hours. The above findings have been presented at a local meeting. A new protocol based on the AEPU algorithm has been adopted for use within the department. We aim to revise our documentation in line with this and repeat the audit process in 6 months time.

A population-based cohort study of uterine carcinosarcoma

M. Gurumruthy¹, R. Lahiri¹, N. Scott², I. Miller¹ & D. E. Parkin¹

¹ Aberdeen Royal Infirmary, ²University of Aberdeen, UK

Aim. To determine the incidence of uterine carcinosarcoma and factors influencing survival.

Design and setting. A retrospective population-based cohort study in Grampian region, Scotland.

Population and methods. Women diagnosed with carcinosarcoma of the uterus residing in the North-east of Scotland were included and analysed in the study. The study period was from 1991– 2009.

Kaplan Meier plots and Cox-regression analysis were used for analysing the results.

Main outcome measures. Overall survival.

Secondary outcome measures. The relationship between incidence, FIGO stage, lymphovascular space invasion, treatment, tamoxifen and 5-year survival rates.

Results. A total of 59 women were diagnosed with carcinosarcoma of the uterus and 43 (72.9%) were included in the analysis. The total number of women with endometrial adenocarcinoma over the same period was 1,642. The median survival was 25 (range) months from the time of confirmed histological diagnosis.

The estimated 5-year survival for Stage I and II was 52.5% (95% CI: 30.5–74.5%) and the 2-year survival rate for Stage III and IV was 46% (95% CI: 16–75%). There was a non-significant increase in incidence in the number of cases over four of the 5-year periods. Improved survival was seen with early stage disease (FIGO Stage I and II) and the absence of lymphovascular space invasion (p = 0.015). Having a heterologous or homologous sarcoma component did not have an effect on the survival outcome. A total of 26% of the women had a history of tamoxifen usage, with no effect demonstrated on survival. Multivariate analysis showed that when treatment modality and lymphovascular space invasion were adjusted for FIGO staging, there was no statistical significance in the survival outcomes.

Conclusion. The small number of cases (3.6% of all uterine tumours) over a 19-year study period reflects the rarity of uterine carcinosarcomas. We confirmed poor prognosis with a median survival of 25 months.

A larger study based on all Health Boards in Scotland could give a more accurate picture of the incidence and behaviour of this tumour, and further our understanding of the efficacy of various treatment modalities.

Management of microinvasive cervical cancer in Tayside (2000–2009)

S. Sinha El-Miligy, W. McMullen & C. McKenzie

Ninewells Hospital, Dundee, UK

Aim. To review the management and outcome of patients presenting with microinvasive disease of the cervix (MIC) in patients diagnosed in Tayside over a period of 10 years (2000–2009). During this time period, a policy of see and treat was common practice with the treatment options being LLETZ or cold coagulation.

Methods. From the cancer database, cases of MIC were identified and a retrospective review of case notes was undertaken. Cross reference was made to subsequent management, histology results and follow-up cytology.

Results. A total of 41 patients were identified, of which three patients were below the age of 25 and two were 25-years-old at the time of diagnosis. Eight patients were between 25 and 30, and 18 were aged 31–40. Only 10 patients were above the age of 40.

In the majority of patients (71%), the referral smear showed severe dyskaryosis. In the other cases it showed moderate dyskaryosis (12%), glandular abnormality (10%) and ? invasive (7%).

At the initial visit to the colposcopy clinic, 29 patients underwent LLETZ (Group A: 70%) and 10 patients underwent punch biopsies and cold coagulation at the same visit (Group B: 24%). In two cases, the colposcopists opted to await the histopathology results of the punch biopsies.

In Group A, 11 patients required follow-up cytology only; nine patients underwent repeat LLETZ; six patients proceeded to hysterectomy and two underwent radical hysterectomy. One patient was lost to follow-up.

In Group B, all 10 patients were invited back for LLETZ, which showed the following features: (1) Negative LLETZ in six cases; (2) CIN-1 in one case, this patient subsequently had negative smears; (3) CIN-2 in one case, subsequent smear was negative; (4) microinvasive disease in two cases; both patients underwent a radical hysterectomy and had negative smears on follow-up. None of the patients who attended for follow-up had recurrence.

Conclusions. This survey reviews the outcome of patients diagnosed with MIC. It is recognised that if MIC is suspected, LLETZ should be undertaken, however, the colposcopic diagnosis of MIC is not always certain. In units who utilise cold coagulation and ‘See and Treat’ management, the possibility exists that patients will be under treated. Over a 10-year period, this occurred in 10 cases. All cases received further treatment and no patient has developed recurrence. In cases where LLETZ (following cold coagulation) has shown no residual disease, it has been difficult to determine the FIGO staging. The MDT discussion has decided the subsequent management – due to the lack of certainty of the FIGO staging this could involve over or under treatment. Although the numbers of cases is small, the unit have reviewed the policy and the current guideline to recommend that if ‘See and Treat’ management is undertaken, treatment must be by LLETZ.

Complications and outcome of minimal descent vaginal hysterectomy vs laparoscopy assisted hysterectomy

B. Pande, P. W. Ashok & M. Mathur

Aberdeen Royal Infirmary, Aberdeen, UK

Aim. To compare significant differences in complications and outcomes between minimal descent vaginal hysterectomy (VH) and laparoscopy-assisted hysterectomy (LH).

Methods. Retrospective study compares 25 cases, each of two types of hysterectomy. Data analysed using SPSS 16. Cases with minimal uterine descent (up to 1st degree) for vaginal hysterectomy were selected (excluding pelvic floor repair). Laparoscopic group included laparoscopy assisted vaginal hysterectomy (LAVH), laparoscopic hysterectomy (LH) and total laparoscopic hysterectomy (TLH).

Results. Mean age for VH was 45 years ± SD = 7, range 29–60 years. Mean age for LH was 43 years ± SD = 7, range 35–69 years. All women in the VH group and 21 (84%) of the LH group were parous. Significant past history included caesarean section in two (8%) of the VH and seven (28%) of the LH group. All women in the VH group had uterine descent, whereas in the LH group, this was seen in five (20%) patients.

Indication was menorrhagia/past failed treatment for menorrhagia in 24 (96%) of VH and 21 (84%) of LH groups. Of the LH group, four patients (16%) were LAVH, one was TLH (4%) but the majority, 20 (80%), were LH. Senior trainees performed 15 (60%) of VH and nine (36%) LH; all procedures were supervised by consultants.

An intraoperative complication was seen in four patients (16%) of VH and one patient (4%) of the LH group (converted to laparotomy). Early postoperative complications were seen in eight patients (32%) of VH and four (16%) of LH. None returned to theatre. Readmission was seen in 24% of VH and 16% of the LH groups. Late complication was equivalent in terms of haematoma (managed conservatively in all). Minor complications included constipation, urine retention in the VH group and infection in the LH groups.

The majority of VH (84%) took 90 min, while 76% of LH took up to 150 min to complete. Of the LH, 88% were discharged within 48 hours, while 80% of the VH went home in 96 hours.

Conclusion. There was in general, safety in both procedures. Hence, in experienced hands with the correct choice of surgery, both modes of management are acceptable in patient care.

VH involves less operating cost in terms of equipment and theatre time. It is worth continuing training towards minimal descent VH.

However, our numbers are too small to make significant conclusions and a larger study would be recommended to evaluate this further.

An audit on the management of heavy menstrual bleeding in NHS primary and secondary care

J. S. Loh

Department of Obstetrics and Gynaecology, Aberdeen Royal Infirmary, Aberdeen, UK

Background. Heavy menstrual bleeding (HMB) is defined as excessive menstrual blood loss occurring alone or in combination with other symptoms, interfering with a woman's physical, emotional, social and material quality of life.

Aim. To compare the management of HMB in primary and secondary care according to the NICE Clinical Guideline 44, which recommends: (a) ultrasound as first-line diagnostic tool for identifying structural abnormality; (b) endometrial biopsy to exclude endometrial cancer or atypical hyperplasia and (c) levonorgestrel-releasing intrauterine system (LNG-IUS) as first-line long-term treatment for HMB.

Methods. Data representing primary care were collected from medical records at Cults Medical Group, a primary care practice equipped with ultrasound and endometrial biopsy. Women aged 30–55 presenting to the practice with HMB from 2007–2009 were targeted. Data representing secondary care were collected from medical records at Aberdeen Royal Infirmary from August to October 2010. The percentage of women receiving: (a) ultrasound; (b) endometrial biopsy and (c) LNG-IUS were identified.

Results. A total of 61 women aged 30–55 presented to primary care with HMB from 2007–2009. However, only 56 women had proper documentation for data collection. A total of 45 women aged 25–55 presented to secondary care with HMB from August to October 2010.

In primary care: (a) 33.93% of women had ultrasound; (b) 25.0% of women had endometrial biopsy and (c) 10.71% of women received LNG-IUS. In secondary care: (a) 100.0% of women had ultrasound; (b) 88.89% of women had endometrial biopsy and (c) 60.0% of women had LNG-IUS.

Conclusion. Secondary care was more compliant with the guidelines than primary care in the management of HMB. Differences in the environment and the accessibility or availability of resources contributed towards a higher compliance in secondary care. It was recommended that primary care should encourage women with HMB to consider LNG-IUS.

Is routine measurement of anti-Müllerian hormone necessary prior to ovarian stimulation in IVF

S. Pandey, M. Hamilton & A. Maheshwari

Aberdeen Maternity Hospital, University of Aberdeen, Aberdeen, UK

Appropriate ovarian stimulation is an important determinant of success in IVF programmes. Various tests of ovarian reserve are described in the literature to guide selection of the optimal gonadotrophin dose. Of all the tests, anti-Müllerian hormone (AMH) seems to be the most popular. This is now used by many units in the UK, its greatest advantage being identification of patients at risk of ovarian hyperstimulation (OHSS). We changed our stimulation protocols in 2010 for individualisation of stimulation regimens and currently use antral follicle count (AFC), age and BMI, all of which are part of routine clinical evaluation. We did not include AMH measurement in our protocols because of costs.

A retrospective audit of all patients undergoing IVF was conducted over a period of 6 months before (January–June 2009) and after (July–December 2010) the full implementation of the new protocol to determine whether our approach, not including AMH assessment, was able to predict and prevent OHSS, while maintaining good success rates. Our standards were: (1) < 1% cancellation due to OHSS prior to egg collection; (2) < 1% freezing of all embryos due to OHSS and (3) no severe OHSS.

A total of 241 (January–June 2009) and 253 (July–December 2010) IVF cycles were initiated. Cancellation prior to egg collection was 0.8% (2009) and 0.3% (2010). All the embryos were frozen due to the risk of OHSS in six cycles in 2009, but only in one cycle in 2010 (p = 0.06). There were no admissions due to severe hyperstimulation in either period. The ongoing pregnancy rate was higher after the change to the individualised protocol (23.7% vs 31.6%, p = 0.048) across all age groups.

Hence, OHSS can be avoided by a structured algorithm, using age, BMI, basal FSH and AFC. AMH was not deemed essential for routine practice, thereby saving over £30,000 a year, without compromising patient care.

Sperm morphology: Should it determine who should have ICSI?

Z. Hasafa, M. Hamilton & A. Maheshwari

Aberdeen Maternity Hospital, Aberdeen, UK

Intracytoplasmic sperm injection (ICSI) was introduced for male factor infertility. ICSI is more invasive and expensive. In addition, long-term outcomes for ICSI are not known, as the oldest ICSI children are only in their late teens. There are concerns that natural sperm selection is lost in carrying out ICSI and the oocyte may be susceptible to damage. Despite evidence of no advantage compared with IVF, ICSI is being widely practiced throughout the world for non-male factor infertility. In some clinics, up to 80% of cycles utilise ICSI, while in our unit, this proportion is only 33%, much lower than national average of 50%. The role of morphology of sperm as an indication for ICSI is controversial and at present, is not included as an indication in our centre. We have audited in our own clinic the effect of morphology on IVF outcome, with the intention of determining a cut-off to make a decision of in whom to do ICSI.

We retrospectively analysed the data from all first IVF cycles in 2010. They were stratified into two groups based on different cut-offs of morphology (> 0%, > 1%, > 2% and 3%) to determine the fertilisation rate at different cut-offs of morphology using standard IVF. The standard rate of fertilisation for IVF should be 60%.

The average fertilisation rate was 67.8% (SD 0.25). There was no statistical difference in fertilisation rate (standard deviation) at various morphologies’ cut-off (0–66%, 29 vs 68%, 24; 1–69%, 25 vs 67%, 25; 2–69%, 25 vs 66%, 25 and 3–70%, 25 vs 64%, 25).

We conclude that there is no indication to include morphology in the criteria for performing ICSI, as it is a poor predictor of fertilisation outcome. Other predictors of fertilisation need to be evaluated to determine suitable cases for ICSI.