Drugs given to pregnant women and their babies to keep them still during fetal surgery and other invasive procedures on the baby

We looked for evidence from randomised controlled trials on the effectiveness of anaesthetic and pain relief medication for preventing fetal movements (fetal immobilisation) to allow surgical and other invasive procedures to be performed safely on the unborn fetus whilst still in the mother's uterus. 

What is the issue?

Use of ultrasound during pregnancy has made great advances resulting in increased diagnosis of problems with the development of unborn babies. Many structural fetal problems as well as complications associated with multiple pregnancies can be treated while babies are still in the mother's uterus. Many problems generally require treatment before birth, for example a blockage in the airways. We set out to determine which medications are effective at providing fetal immobilisation for in-utero procedures to be performed, as well as looking at the effect on the mother in terms of sedation and pain relief. For these procedures to be performed safely, fetal movements may need to be reduced. This can be done by giving medication either to the mother (usually by injection into a vein) or directly to the fetus (by injecting into fetal muscle).

Why is this important?

Treatment of structural issues such as lung or heart problems can greatly improve pregnancy and newborn outcomes. Similarly, with multiple pregnancy complications such as twin-twin transfusion syndrome, where there is unequal sharing of the one placenta between the twins. Providing the best surgical conditions with reduced fetal movements means that the procedure can be performed safely and reduces complications such as preterm birth. It is also important to provide a safe environment for the mother without side effects such as oversedation or lack of pain relief during the procedure.

What evidence did we find?

We searched for evidence from randomised controlled trials to May 2021 and identified one eligible trial involving 54 pregnant women. All were women with a multiple pregnancy in their second trimester attending a European hospital for specialist care. The women received intravenous injection of either remifentanil or diazepam to immobilise the fetus and to sedate the mother during fetoscopic surgery. This is where a small instrument (laparoscope) is inserted through the mother's abdominal wall and into the uterus to see the fetus and placenta. The trial gave low-certainty evidence. The main outcome looked at was how much the baby moved via assessment of body and limb movements. Maternal outcomes were the amount of sedation and depression of breathing. The time for surgery and surgical satisfaction with the procedural conditions were also measured. Low-certainty evidence suggested that remifentanil may be better at reducing fetal movements when compared to diazepam and surgeons were more satisfied with the procedure when using remifentanil.  However, the amount of maternal sedation and depression of breathing were worse with remifentanil. 

What does this mean?

Low-certainty evidence from one small trial means that we cannot be confident in the results. The information available on maternal, fetal, and newborn outcomes was limited. More high-quality trials are needed to increase our certainty about the longer term fetal and newborn impact of the medication used for fetal immobilisation for in-utero procedures, as well as other medications which may be suitable for use.


Authors' conclusions: 

We were only able to include one study with a small number of women, from a single centre, a European tertiary hospital. This study was published in 2005 with an abstract of this trial published in 2004. This study evaluated two intravenous medications administered to the mother - remifentanil and diazepam. This study reported our prespecified primary outcome but only evaluated several of our secondary outcomes, which limited further assessment. Low-certainty evidence suggested that remifentanil may be better at reducing fetal movements and surgeons were more satisfied with the procedure. However, maternal sedation and depression of breathing may be worse with remifentanil. 

Further high-quality RCTs assessing both fetal and maternal medications are required to evaluate their efficacy for fetal immobilisation as well as safety for both mother and fetus.

Read the full abstract...

Developments in ultrasound assessment of pregnancy has resulted in the increasing diagnosis of antenatal fetal issues. Many structural fetal conditions as well as complications associated with multiple pregnancies have the potential for in-utero treatment to improve both pregnancy and neonatal outcomes. Procedures such as laser ablation for twin-twin syndrome or cord occlusion for selective fetal termination require fetal immobilisation. Immobilisation of the fetus can occur through administration of medication to the mother or directly to the fetus. This improves procedural success and reduces the ongoing risk to the pregnancy. Evidence regarding the best medication and mode of delivery helps to ensure the optimal decision is made for both the mother and the fetus.


To assess the effects of perioperative pharmacological interventions for fetal immobilisation during fetal surgery and invasive procedures on fetal, neonatal, and maternal outcomes. 

Search strategy: 

We searched Cochrane Pregnancy and Childbirth’s Trials Register, ClinicalTrials.gov, the WHO International Clinical Trials Registry Platform (ICTRP) (10 May 2021), and reference lists of retrieved studies.

Selection criteria: 

We included randomised controlled trials (RCTs) and quasi-RCTs (including published abstracts) which compared different classes of medication administered to the mother or fetus to allow in-utero procedures to be performed. We also included cluster-randomised trials but excluded cross-over trials.

Data collection and analysis: 

We used the standard Cochrane Pregnancy and Childbirth methods for data collection and analysis. Two review authors independently assessed trials for inclusion and risk of bias, extracted data, and checked them for accuracy.

Main results: 

One study with three trial reports met the inclusion criteria. This involved 54 women with a multiple pregnancy. The study was conducted in a tertiary European hospital maternal-fetal medicine unit and compared remifentanil to diazepam for fetal immobilisation and maternal sedation during fetoscopic surgery.  

Low-certainty evidence suggested that remifentanil may reduce fetal movement more than diazepam for two outcomes of fetal movement, one of fetal immobilisation at 40 minutes using a visual analogue score (VAS) (where 0 = immobile and 100 = baseline mobility), and one of gross body and limb movements (score was absolute number of movements), both assessed by a sonographer evaluating a taped ultrasound sequence (mean difference (MD) -65.00, 95% confidence interval (CI) -69.38 to -60.62 and MD -10.00, 95% CI -11.62 to -8.38; 1 study, 50 women). 

Surgeons may also report being more satisfied with the procedure when using remifentanil rather than diazepam (risk ratio (RR) 2.88, 95% CI 1.60 to 5.15; 1 study, 50 women; low-certainty evidence). However, maternal respiratory rate may decrease more during the surgical procedure with remifentanil compared with diazepam (MD -6.00, 95% CI -8.29 to -3.71; 1 study, 50 women; low-certainty evidence). Maternal sedation may also be worse with remifentanil compared with diazepam (RR 0.09, 95% CI 0.01 to 0.65; 1 study, 50 women; low-certainty evidence) measured using an observer assessment of alertness/sedation (where a score of < 4 equates to profound sedation and > 4 equates to insufficient sedation). Perinatal mortality and time taken to perform the procedure were not reported in the trial. 

We prespecified 20 outcomes and planned to use GRADE for 6 of them, all other outcomes were not able to be reported against for the purpose of meta-analysis due to data not being provided or unable to be interpreted.

We assessed the included study at low risk of selection bias (appropriate random sequence generation and allocation concealment), performance bias (blinding of participants and personnel), detection bias (outcome assessors were blinded), attrition bias (incomplete outcome data minimal), and reporting bias. 

Our GRADE assessment for certainty of the evidence indicates that there is low certainty of the evidence.