Interruption of a pregnancy after 14 weeks gestation may be required when the fetus is dead, severely malformed or in cases of maternal illness. This process is usually conducted medically in Australia, using the prostaglandin E1 analogue misoprostol. This prostaglandin, although not specifically licensed for use in pregnancy termination, is now a common abortifacient with a lot of accumulated experience both within Australia and internationally.

Since 1996, misoprostol, a synthetic prostaglandin, has been used at King Edward Memorial Hospital as the principal agent for second trimester pregnancy termination. This agent is administered vaginally, and in its current form and dosage regimen results in 75-80% of women delivering within 24 hours. As experience with this agent has grown, it has been observed that in approximately 40% of women the placenta is either completely retained or incompletely delivered, necessitating operative removal and an increased potential for maternal blood loss. In this study, it is planned, in a randomized controlled clinical trial, to evaluate three regimens for the management of placental delivery in women undergoing second trimester pregnancy interruption. The primary intention of this study is to develop a third stage management protocol to reduce the incidence of placental retention in second trimester medical pregnancy termination.

The secondary aim of this study is to assess the ultrasound appearance of the uterus and its cavity within 24 hours of second trimester pregnancy termination. The ultrasound appearances of the uterus following second trimester pregnancy loss have not been previously investigated in detail. Previous ultrasound studies of the term postpartum uterus have demonstrated a high incidence of echogenic material within the uterine cavity soon after an uncomplicated vaginal delivery. These findings have been of concern as the ultrasound appearances may erroneously imply a need for operative intervention. The investigators wish to ascertain if this high incidence of echogenic tissue presence is also true in the second trimester. Ultrasound is frequently used by clinicians to define placental completeness and the potential requirement for surgical curettage. The data from this single sonographic examination of the uterus will provide baseline data for a planned longitudinal study of uterine appearances following second trimester pregnancy loss and their correlation with clinical symptoms.

Methods:

All women who are admitted to King Edward Memorial Hospital for Women for pregnancy interruption for severe fetal anomaly or maternal pregnancy complications between 14 and 24 weeks gestation will be invited to participate in the study. No women with an intrauterine fetal demise will be recruited due to the potential confounding effects of the fetal death on the placental separation process.

Once consent has been obtained, the women will be randomised to three placental management strategy groups:

Group 1: Standard management protocol; Group 2: Oxytocin protocol; Group 3: Oral misoprostol protocol.

Group 1:

Women allocated to the standard management protocol will receive no routine oxytocic following the delivery of the fetus.

If spontaneous placental expulsion has not occurred within 60 minutes of delivery, or if heavy vaginal bleeding ensues within that time period, manual removal of the placenta will be performed in the operating room under general or regional anaesthesia. No cord traction is used to facilitate placental expulsion, although maternal expulsive efforts or digital extraction if the placenta is visible at the vaginal introitus are permissible.

Group 2:

Women allocated to the oxytocin protocol will receive a single intramuscular injection (IMI) into the upper thigh of 10 IU oxytocin (Syntocinon®) as soon as the fetus is expelled.

If placental expulsion has not occurred within 60 minutes of delivery, or if heavy vaginal bleeding ensues within that time period, manual removal of the placenta will be performed in the operating room under general or regional anaesthesia. No cord traction is used to facilitate placental expulsion, although maternal expulsive efforts or digital extraction if the placenta is visible at the vaginal introitus are permissible.

Group 3:

Women allocated to the oral misoprostol protocol will receive a single oral dose of 600 mcg misoprostol as soon as the fetus is expelled.

If placental expulsion has not occurred within 60 minutes of delivery, or if heavy vaginal bleeding ensues within that time period, manual removal of the placenta will be performed in the operating room under general or regional anaesthesia. No cord traction is used to facilitate placental expulsion, although maternal expulsive efforts or digital extraction if the placenta is visible at the vaginal introitus are permissible.

The midwife or nurse caring for the woman will inspect all placentas visually after non-surgical expulsion for completeness. This may not occur in the case of obvious incomplete expulsion of the placenta, as the woman will be transferred to the operating room for completion of evacuation on the basis of clinical need. Due to the placental fragmentation that may occur during manual removal of the placenta in the operating room it will not be possible for those women who require surgical removal of their placenta to have a visual placental inspection, however the surgeon will assess the completeness of the removal digitally in the operating room.

A maternal full blood count will be performed prior to the commencement of the termination process and repeated 6 hours after the delivery of the placenta, as an index of blood loss. The pre-termination full blood count is a standard investigation in the current clinical pathway. The attending nurse or midwife will weigh all pads/sheets after fetal delivery until placental expulsion to estimate blood loss in the three groups.

Maternal pulse, blood pressure and temperature will be recorded every 15 minutes from fetal expulsion until 1 hour post-placental delivery.

Maternal symptoms of nausea, headache and abdominal pain will be recorded every 15 minutes on a visual analogue scale from fetal expulsion until placental delivery or transfer to the operating room. Maternal emesis will be recorded from fetal expulsion until 1-hour post-placental delivery.

All women will undergo a transabdominal ultrasound assessment of the uterine cavity prior to hospital discharge (approximately 6-12 hours post-delivery) in the King Edward Memorial Hospital Diagnostic Imaging Department. The uterine size, myometrial thickness, myometrial vascularity, uterine cavity dimensions and appearance of the uterine cavity contents will be assessed sonographically. Myometrial vascularity will be correlated with the pre-delivery placental position. All images will be reviewed by the Chief Investigator, who will be blind to clinical outcomes, using the hospital PACS. The Chief Investigator (who has formal ultrasound qualifications) will assess the image quality, measurements and appearances of the uterine cavity. The ultrasound appearances of those women who have undergone an operative placental removal will be used as the baseline control group for uterine cavity appearances, as it is assumed there is no placental tissue present in this group. The ultrasound appearances of the uterus will not be reported to the clinicians caring for the woman, as the evidence from the term uterine cavity ultrasound studies is that echogenic material is frequent and does not impact adversely upon the postnatal course. The exception to this will be when the clinician caring for the woman requests an ultrasound evaluation of the uterine cavity for suspected retained placental tissue.

It is estimated that 3-5 women per week will be eligible for recruitment into this study and that this additional workload can be conducted without adversely impacting on other duties of the Diagnostic Imaging Department. The Chief Investigator is the head of the KEMH Ultrasound Department and is responsible for workload distribution within the unit.

All women will be telephoned at 4 weeks post-termination by the Chief Investigator to enquire about post-discharge bleeding complications (eg. antibiotic prescription, curettage, duration of vaginal bleeding).

The primary objectives to be achieved from the conduct of this study are:

1. The efficacy of routine ecbolic agents to facilitate spontaneous placental expulsion compared with a non-pharmacologic intervention. This will primarily be determined by ascertaining the percentage of women in each group who require operative removal of the placenta.

   The investigators' aim is to reduce the requirement for operative placental removal from the current 40% to 20%.

2. Comparison of the frequency and severity of side-effects associated with routine ecbolic administration in the third stage.

3. The sonographic appearances of the uterus and its cavity in the immediate post-delivery period. It is planned in future studies to assess the sonographic appearances of the second trimester uterus after delivery in a longitudinal manner over time, using these data as baseline information.

Sample size:

Up to 300 women will be recruited into the study (100 per group). This sample size will achieve 80% power to detect a reduction from 40% placental retention on the current regimen of non-pharmacological third stage management to 20% placental retention using either 5IU oxytocin IMI after delivery or 600 mcg misoprostol orally. Two pairwise comparisons between the current regimen and a new regimen will be conducted at 0.025 significance level each to attain an overall 0.05 significance level using a test of proportions. This sample size also allows for an interim analysis consisting of two pairwise tests, when 60 patients per group have been recruited (at the nominal significance level of 0.003 per test, critical Z-statistic=3.02) and a final analysis (at the nominal significance level of 0.024 per test, critical Z-statistic=2.25). Sequential sample size calculations were obtained using O'Brien-Fleming spending function for the test statistics boundaries (PASS 2002 for Windows, Kaysville Utah). At the time of interim analysis, the placental retention rate per study arm will be estimated and a possibility of stopping recruitment to one of the arms due to futility will be considered.

Statistical Analysis:

Descriptive statistics will utilise means and standard deviations (or medians and interquartile ranges) for continuous data and frequency distributions for categorical data. The primary endpoint will be evaluated using pairwise tests of proportions. Secondary analyses will implement logistic regression modelling to investigate other patient characteristics that are relevant for prediction of placental retention.

Inclusion Criteria:

• Abortion at 14-24 weeks gestation
• Live fetus
• Medical termination with vaginal misoprostol

Exclusion Criteria:

• Surgical termination
• Gestation less than 14 weeks or greater than 24 weeks

Planned interim analysis not performed as mifepristone became available in Australia. Study ceased at 251 women and enough power available to calculate the groups as a single stage analysis.