Multifetal pregnancy reduction and selective termination

INTRODUCTION — Multifetal gestations are at higher risk for a variety of maternal, fetal, and neonatal complications than singleton pregnancies. Ideally, multiple gestations should be prevented by better use of assisted reproductive technology (eg, better management of ovulation induction and embryo transfer). In the absence or failure of preventive interventions, or in the setting of a spontaneous conception, procedures for multifetal pregnancy reduction (MPR) have been developed to reduce the overall number of fetuses in the gestation and thereby improve maternal outcome and the outcomes of the surviving fetus(es). The fetus(es) chosen for reduction is mostly chosen randomly, but with consideration of technical factors.

Selective termination (ST) also involves reduction of the fetal number, but the ST procedure is directed at one or more specific fetuses in a multiple gestation because of a known genetic, structural, or other abnormality (eg, severe growth restriction) that has been identified by ultrasound examination or by an invasive procedure (amniocentesis, chorionic villus biopsy) for fetal diagnostic testing.

This topic will discuss MPR and ST. Additional aspects of management of multifetal gestations and strategies to reduce the occurrence of twin and higher order pregnancies are reviewed separately.

●(See "Twin pregnancy: Overview".)

●(See "Twin pregnancy: Labor and delivery".)

●(See "Triplet pregnancy".)

●(See "Strategies to control the rate of high order multiple gestation".)

In this topic, when discussing study results, we will use the terms "woman/en" or "patient(s)" as they are used in the studies presented. However, we encourage the reader to consider the specific counseling and treatment needs of transgender and gender diverse individuals.

MULTIFETAL PREGNANCY REDUCTION

Definition — MPR refers to a procedure in which the overall number of fetuses in a multifetal gestation is reduced by terminating one or more fetuses mostly chosen randomly, but with consideration of technical factors.

Rationale — It is well-established that the morbidity and mortality of multifetal gestations increase with increasing numbers of fetuses and that adverse outcomes in offspring are primarily the consequence of preterm birth [1-5] (see UpToDate topic reviews on multiple gestation and on the preterm infant). The primary justification for MPR is that the "take home" baby rate per pregnancy is increased by increasing the gestational age at birth and birth weight, thus reducing morbidity and mortality from preterm birth [6]. The risks for some maternal complications, such as preeclampsia, are also reduced [7].

Improved medical outcomes are not the only consideration. The economic and psychological impacts of multiple gestation on families are also important and can be additional reasons for MPR [8,9]. The emotional stress and complexity of caring for multiple newborns, especially those with medical or developmental issues, is high. Over 25 percent of parents of multiples experience depression or anxiety disorders in the perinatal period [10]. In addition to direct medical costs, infants/children with complex medical and developmental issues often require specialized and coordinated care, preventing parents and family members from returning to work and thus increasing financial stress.

In response to these and other issues, an American College of Obstetricians and Gynecologists (ACOG) Committee Opinion on multifetal pregnancy reduction states that "obstetrician-gynecologists should respect patients' autonomy regarding whether to continue or reduce a multifetal pregnancy" and that "only the patient can weigh the relative importance of the medical, ethical, religious, and socioeconomic factors and determine the best course of action for her unique situation" [11].

Candidates

●Multichorionic triplet and higher order pregnancies – MPR should be offered for any pregnancy with three or more fetuses, as there is good evidence from meta-analyses of observational studies that it improves neonatal survival rates and outcomes in survivors [12,13].

An International Federation of Gynecology and Obstetrics (FIGO) Committee Report stated that "multiple pregnancy of an order of magnitude higher than twins involves great danger for the woman's health and also for her fetuses, which are likely to be delivered prematurely with a high risk of either dying or suffering damage" and "where such pregnancies arise, it may be considered ethically preferable to reduce the number of fetuses rather than to do nothing" [14].

Outcome data of electively reduced versus ongoing triplet pregnancies are discussed below. (See 'Triplets to twins' below.)

●Dichorionic diamniotic twin pregnancies – MPR can be considered for dichorionic diamniotic (DCDA) twin pregnancies, particularly in the setting of a medical or obstetric indication, such as maternal medical disease (eg, cardiac disease, renal disease) or history of preterm delivery or cervical insufficiency. Reduction to a singleton gestation can also be considered in women of very advanced maternal age (≥45 years old), as they are at high risk for pregnancy complications, such as preterm delivery, preeclampsia, growth restriction, and possibly gestational diabetes with a twin gestation, as compared with a singleton gestation [15].

MPR in the absence of a medical/obstetric indication is controversial, given that the outcome of twin pregnancies is much less dire than for higher order gestations [11,16-19]. However, even in the absence of medical or obstetric indications for MPR, some patients feel that they can meet the needs of one additional child but not two additional children. These patients should be counseled objectively regarding their options [11].

Outcome data of electively reduced versus ongoing twin pregnancies are discussed below. (See 'Twins to singletons' below.)

●Selected monochorionic gestations – The injection techniques discussed below for MPR are intended for dichorionic fetuses or for reduction of all of the fetuses in a monochorionic gestation (such as a monochorionic pair of a triplet or higher order gestation).

Reduction of a single fetus in a monochorionic gestation requires specialized cord occlusion techniques to avoid death or neurologic harm to the co-fetus from effects resulting from placental vascular anastomoses. These cord occlusion techniques are associated with a higher risk of miscarriage than the injection technique described below and, therefore, are rarely performed for MPR. On those rare occasions when a patient chooses to electively terminate a single fetus of a monochorionic gestation, the technique is the same as that described below for ST of a monochorionic gestation. (See 'Monochorionic fetuses' below.)

Contraindications to multifetal pregnancy reduction — There are no absolute contraindications to MPR, but some relative contraindications exist.

The inability to confidently diagnose chorionicity, although uncommon, is a relative contraindication to MPR because the demise of one fetus of a monochorionic multiple gestation can be harmful to the coexisting fetuses in the monochorionic sac (see 'Monochorionic fetuses' below). HIV or active hepatitis is another consideration when planning an invasive in utero procedure, due to the risk of vertical transmission; however, the actual risk is unknown and is probably low. Conditions that make the procedure overly technically challenging, such as maternal obesity or impaired sonographic visualization of the fetus due to previous abdominal surgery, can also be relative contraindications, although most of these technical considerations can be overcome in experienced hands.

Cervical shortening, a relative contraindication for many other in utero procedures, is not necessarily considered a contraindication for MPR in our practice.

Preprocedure considerations

●Informed consent and counseling – Patients should have a clear understanding of the risks of multiple gestation and the benefits and risks of MPR. The primary risk associated with MPR is miscarriage, but this risk should be interpreted in the context of miscarriage of expectantly managed multiple gestations, as the risk of miscarriage after MPR is lower than that for spontaneous loss of unreduced triplets or higher order multiples [6]. (See 'Pregnancy outcomes' below.)

●Local regulations – MPR is philosophically different from pregnancy termination, as it is a procedure that was developed to increase the likelihood of a successful pregnancy, and, as such, we do not consider it a type of abortion. However, since laws vary from state to state and among countries, it is important to be familiar with local legal requirements, such as a waiting period between signing the consent form and undergoing the procedure or gestational age restrictions.

●Chorioamnionicity – Chorioamnionicity is best established by ultrasound examination in the first trimester. (See "Twin pregnancy: Overview", section on 'Assessment of chorionicity and amnionicity'.)

This information is crucial because of potential consequences to the co-twin, including neurologic injury and demise, when interfetal placental vascular anastomoses are present in monochorionic multiple gestations. Cardioplegia with intracardiac potassium chloride injection, which is commonly used for MPR in multichorionic multiples, is not appropriate in monochorionic multiples, which require a different technique. (See 'Monochorionic fetuses' below.)

●Scheduling – MPR is usually performed between 10+0 and 13+6 weeks of gestation, although it can be performed at any gestational age at which legal termination of pregnancy is possible. Reduction at this early gestational age allows parents to undergo the procedure while the pregnancy is relatively private, allows time for most spontaneous fetal losses to occur, and allows for at least some assessment of fetal health. Options for fetal assessment prior to MPR, as discussed in the bullet below, should be reviewed and the MPR scheduled accordingly.

The risk of procedure-related pregnancy loss is thought to be low at this time. There may be a slight increase in risk of loss at 6 to 8 weeks [20,21] and after 16 to 18 weeks of gestation [22], but data are discordant.

●Fetal assessment – A survey of fetal anatomy is performed before MPR to look for structural anomalies and markers of aneuploidy, such as increased nuchal translucency or absent nasal bone. A detailed first-trimester anatomic assessment at 12+0 and 13+6 weeks will maximize the assessment, but even prior to 12 weeks, many fetal abnormalities will be detected [23]. Forty percent of lethal anomalies can be detected by the end of the first trimester; however, this early assessment should not replace a comprehensive anatomic survey at 18 to 22 weeks of gestation, which will detect more anomalies [24]. As evaluation of fetuses in the first trimester is limited, patients should be clearly counseled that the ongoing health of nonreduced fetuses cannot be guaranteed. Although rare, we have been involved in cases in which nonreduced fetuses are diagnosed with congenital or acquired abnormalities later in gestation. (See "Sonographic findings associated with fetal aneuploidy".)

Chorionic villus sampling (CVS) can be performed on one or more fetuses prior to MPR. Performing CVS before MPR does not appear to increase the postprocedural pregnancy loss rate above that of MPR alone and may decrease losses [25-28]. CVS may be performed on all fetuses or only on the fetus(es) not planned for reduction. At our center, the majority of patients opt to have a full karyotype and microarray before MPR; results are available in approximately 7 to 14 days. In rare cases, patients who reside distant from our center undergo MPR after fluorescent in situ hybridization (FISH) analysis for chromosomes 13, 18, 21, X, and Y alone, which provides results within 24 to 48 hours. However, this is not recommended to be routine practice since 0.3 percent of clinically significant unbalanced chromosomal rearrangements will be missed with CVS for FISH alone, approximately 45 percent of which may be associated with a clinically significant phenotype [29,30]. Microarray will further identify clinically relevant deletions or duplications in 6.0 percent of fetuses with a structural abnormality and up to 1.7 percent of fetuses with a normal ultrasound [31].

Increasing use of preimplantation genetic testing for aneuploidy (PGT-A) has prompted controversy about the utility of performing CVS before or after MPR, but PGT-A may not be definitive, so confirmatory testing with CVS is still recommended [32]. (See "Chorionic villus sampling" and "Preimplantation genetic testing".)

At the time of CVS, an accurate description, as well as a diagram of the absolute and relative positions of the gestational sacs and placentas, must be carefully recorded so that if an abnormal karyotype is identified, it can be correctly matched to the affected fetus, as many karyotypically abnormal fetuses will be indistinguishable from normal fetuses at this gestational age. Attention should also be paid to the relative fullness of the maternal bladder, as filling and emptying the bladder can change the apparent location of the sacs/placentas. Diagnostic error due to incorrect matching of genetic test results with the fetus or sampling error has been reported in 0.6 to 0.8 percent of twins and up to 1.2 percent of higher order multiple gestations [27,33-36].

●Ending number for triplet and higher order multiples – Most higher order multifetal pregnancies are reduced to twins since many patients desire a twin rather than a singleton birth after years of infertility [37-40]. However, reduction of triplet and higher order multiple gestations to a singleton is becoming more common and associated with good outcomes [41-44]. More data are needed to determine whether reduction of these pregnancies to a singleton significantly improves pregnancy outcome over a reduction to twins, which typically has good pregnancy outcomes compared with higher order multifetal gestations. (See 'Pregnancy outcomes' below.)

Technique — The fetus(es) chosen for MPR is generally closest to the anterior uterine wall and/or the fundus, as these sites are most accessible to needle insertion. However, if a fetus has a lagging crown rump length, a substantially smaller sac, markers suggestive of aneuploidy, or an obvious anomaly, then that fetus is preferentially chosen for MPR since these findings are associated with adverse pregnancy outcome (eg, miscarriage, fetal aneuploidy).

The most common technique uses a transabdominal (TA) approach. A transcervical or a transvaginal approach can be used, but these have been associated with high postprocedural pregnancy loss rates when compared with the TA approach; therefore, these approaches are usually reserved for situations in which the TA approach is not technically feasible [45]. The majority of MPR can be done by the TA approach. (See 'Pregnancy outcomes' below.)

Antibiotic prophylaxis is not routinely recommended as no benefit has been established.

The procedure is most commonly performed by injecting approximately 2 to 3 mL of potassium chloride (concentration 2 mEq KCl/mL) into the fetal thoracic cavity using a 22-gauge spinal needle under ultrasound guidance. Asystole is usually seen within one minute of injection, and total procedural time is typically less than five minutes. Additional fetuses can be reduced with the same needle puncture or, more commonly, with a separate needle puncture. Lidocaine, digoxin, hyperosmolar urea, and cardiac tamponade with saline have also be used successfully to induce asystole, although they are more commonly used at later gestational ages [46].

We typically perform an ultrasound examination one hour after the procedure to confirm asystole in the reduced fetus(es) and cardiac activity in the nonreduced fetus(es). If fetal cardiac activity has resumed in the reduced fetus(es), which is rare, the procedure is repeated.

Postprocedure considerations and follow-up — A small amount of cramping or leaking of amniotic fluid from the sac(s) of the reduced fetus(es) may occur. Vaginal bleeding is uncommon, so patients with bleeding should be evaluated for possible causes.

We perform an ultrasound examination to confirm fetal well-being of the remaining fetus(es) one to two weeks after the procedure. The reduced fetus(es) typically becomes desiccated and compressed over time and becomes incorporated into the fetal membranes/placenta of the nonreduced fetus(es).

Laboratory changes following MPR and ST are described below. (See 'Postprocedure laboratory changes (MPR or ST)' below.)

We perform monthly ultrasounds for fetal growth in the third trimester. We have observed lower birth weights in singletons reduced from triplets or above than in singletons reduced from twins (2489 to 2897 grams [in singletons from triplets or more] versus 3122 grams [in singletons from twins]) [41].

Prenatal care is otherwise routine and depends on the number of remaining fetuses.

●(See "Prenatal care: Initial assessment" and "Prenatal care: Second and third trimesters".)

●(See "Twin pregnancy: Overview" and "Twin pregnancy: Labor and delivery".)

●(See "Triplet pregnancy".)

Antenatal fetal testing (eg, nonstress test) is not indicated for otherwise uncomplicated pregnancies; it should be reserved for routine obstetric indications. (See "Overview of antepartum fetal assessment".)

MPR does not affect decision making regarding mode of delivery; however, whenever possible, delivering providers should be aware of the history of MPR to minimize the risk of retained gestational material. Although reduced fetuses/placentas are usually delivered with the placenta(s) of nonreduced fetus(es), there is a small risk of retention of gestational material. The placenta should be examined carefully for the presence of a reduced fetus/sac, which can be identified in most cases.

Psychosocial outcomes — The decision to undergo MPR, usually after many anxious years of infertility, is a difficult one for most patients. Being selective when sharing information about fertility and infertility treatments, pregnancy, and MPR with family, friends, and acquaintances can help reduce stress associated with these issues [47,48].

Long-term psychosocial outcomes were illustrated in a prospective study that compared patients undergoing MPR over a two-year period with those who did not reduce their triplet pregnancy [49]. Although many in the MPR group expressed sadness and guilt one year after birth of the remaining fetuses, the majority overcame the pain associated with the reduction at two years. Furthermore, when compared with mothers of nonreduced triplets, mothers who had undergone MPR had less anxiety and depression and a more satisfactory relationship with their children [49]. (See "Neonatal complications of multiple births", section on 'Family/caregiver support'.)

Pregnancy outcomes

Summary — No randomized trials of MPR have been performed, as randomization in this setting would not be acceptable to most couples and may not be ethically appropriate [50]. Data from observational studies show consistently that pregnancy outcomes after MPR, particularly preterm birth rates, improve significantly. Outcomes also improve with increased operator experience [19,42].

Triplets to twins — MPR from triplet to twin gestation is associated with later gestational age at birth and reduction in obstetric complications [13,51,52]. Triplet reduction to singleton pregnancy, compared with reduction to twin pregnancy, may be more advantageous as reduction to singleton pregnancy is associated with greater gestational age at birth and improved survival, although supporting data are limited [52,53].

●A 2017 meta-analysis including 24 studies comparing triplet pregnancies reduced to twins with nonreduced (expectantly managed) triplet pregnancies provided useful data for counseling patients with triplets on pregnancy outcomes of MPR [13]:

•Preterm birth – Reduction to twins was associated with:

-Delivery at a later gestational age (mean difference 2.61 weeks, 95% CI 2.09-3.12)

-A 60 to 70 percent reduction in early preterm delivery (<32 weeks of gestation: odds ratio [OR] 0.30, 95% CI 0.22-0.41; <28 weeks of gestation: OR 0.40, 95% CI 0.22-0.71)

-A reduction in neonatal mortality (OR 0.36, 95% CI 0.15-0.83)

-Higher birth weight (mean difference 500 g, 95% CI 439.95-560.04)

•Other adverse pregnancy outcomes were also improved in the reduced group:

-Gestational diabetes (OR 0.35, 95% CI 0.19-0.67)

-Hypertensive disorder of pregnancy (OR 0.47, 95% CI 0.31-0.72)

-Need for prenatal hospitalization (OR -10.46, 95% CI -17.12 to -3.81)

-Cesarean delivery (OR 0.18, 95% CI 0.10-0.33)

•Pregnancy loss – The risk of pregnancy loss <24 weeks of gestation was similar in both groups (7.4 versus 8.1 percent, OR 0.87, 95% CI 0.52-1.42).

●Another 2017 meta-analysis separately assessed outcomes of ongoing trichorionic triamniotic triplet pregnancies (TCTA) and dichorionic triamniotic triplet pregnancies (DCTA) versus those electively reduced to twins [12].

•In TCTA pregnancies, reduction was associated with a substantial reduction in preterm birth <34 weeks (17.3 versus 50.2 percent, relative risk [RR] 0.36, 95% CI 0.28-0.48), without a substantial increase in the rate of miscarriage (8.1 versus 7.4 percent, RR 1.08, 95% CI 0.58-1.98).

•In DCTA pregnancies, ongoing pregnancies had a high rate of preterm birth <34 weeks (51.9 percent) that was only modestly reduced (to 46.2 percent) with elective reduction of the triplet with the separate placenta, but there were too few of these cases to allow confident conclusions about the utility of the procedure in DCTA.

Twins to singletons — Reduction to a singleton from twins has been associated with improved pregnancy outcomes, particularly when the reduction is performed prior to 14 weeks gestational age [54]. Supporting data include the following:

●A 2020 systematic review and meta-analysis of six studies involving 7398 participants showed that MPR of dichorionic twins to singletons was associated with a 30 percent lower risk of preterm birth (five studies, RR 0.30, 95% CI 0.22-0.40, p<0.001) and higher birth weight (four studies, mean differences 548.1 grams, 95% CI 424.04-672.15, p<0.001) than expectant management [55]. They also found no difference in miscarriages between the reduced and expectantly managed participants.

●In a 2019 observational study comparing 250 dichorionic diamniotic twins undergoing elective reduction to a singleton gestation with 605 ongoing dichorionic diamniotic twins, patients who elected to reduce to a singleton pregnancy experienced higher gestational age at delivery and lower rates of preterm birth and pregnancy complications without an increased risk of pregnancy loss. Specific data included (adjusted odds ratio [aOR]) [56]:

•Preterm birth less than 37 and 34 weeks gestation (aOR 5.26, 95% CI 3.67-8.61 and 2.22, 95% CI 1.20-4.11, respectively)

•Cesarean birth (aOR 5.53, 95% CI 3.60-8.49)

•Preeclampsia (aOR 3.33, 95% CI 1.60-6.96)

•Preterm premature rupture of membranes (aOR 3.86, 95% CI 2.00-7.430

Similar outcomes were reported between the groups for total pregnancy loss (at both 20 and 24 weeks), unintended pregnancy loss, intrauterine growth restriction, placental abruption, and gestational diabetes. In this study, patients who elected to reduce to a singleton pregnancy had a higher gestational age of delivery and lower rates of preterm birth and pregnancy complications without an increased risk of pregnancy loss.

●In a retrospective review comparing pregnancy outcomes of twin pregnancies reduced to singletons (n = 250) with ongoing (nonreduced) twin pregnancies (n = 605) [56], MPR was associated with:

•Higher mean gestational age at delivery (39 versus 36.7 weeks)

•Reduction in preterm birth <37 weeks (18 versus 54 percent) and <34 weeks (7 versus 17 percent)

•Reduction in birth weight <10^th percentile (15 versus 38 percent)

•Reduction in birth weight <5^th percentile (9 versus 22 percent)

•Reduction in cesarean delivery (47 versus 79 percent)

•Reduction in preeclampsia (4 versus 17 percent)

•Reduction in preterm prelabor rupture of membranes (6 versus 19 percent)

•Reduction in gestational diabetes (5 versus 9 percent)

•No significant increase in unintended pregnancy loss <24 weeks (2.4 versus 2.3 percent)

Higher order pregnancies reduced to twins versus to singletons — In our single-center observational series of over 840 MPRs with known outcome, the pregnancy loss rate for patients reducing from a multiple gestation (2 to 5+) to twins was higher than for reduction to singleton (5.3 versus 3.8 percent) [41]. Specifically, loss rates for quintuplets, quadruplets, and triplets to twins were 12.1, 5.8, and 4.5 percent, respectively. Loss rates for quintuplets, quadruplets, triplets, and twins to a singleton were 0, 4.0, 6.1, and 2.1 percent, respectively.

The mean gestational age at delivery was inversely proportional to finishing numbers: the mean gestational age of delivery of singletons, twins, and triplets was 38.0, 35.2, and 30.3 weeks.

SELECTIVE TERMINATION

Definition — ST refers to a procedure in which one or more specific fetuses of a multifetal gestation are terminated due to a known or suspected fetal abnormality, including a chromosomal, structural, genetic, or growth abnormality, which has been identified by ultrasound examination or by definitive fetal diagnostic testing after amniocentesis or chorionic villus sampling. This is in contrast to MPR, in which the fetus(es) is chosen randomly, but with consideration of technical factors.

Rationale — ST was initially offered as an option to parents of fetuses with major nonlethal anomalies that could lead to the live birth and long-term survival of a severely impaired child. ST is now also offered to parents of fetuses with lethal anomalies because, for some parents, it is emotionally preferable to undergo ST than to continue to carry a fetus that will die before or shortly after birth. In addition, it is theorized that, in some cases, termination of the anomalous sibling may optimize the outcome of the normal sibling(s). For example, anencephaly, a lethal anomaly, is often complicated by severe polyhydramnios due to impaired fetal swallowing, which increases the risk for preterm labor. Termination of the anencephalic fetus, and subsequent resolution of the polyhydramnios, theoretically may decrease the risk for preterm labor and improve outcome for the co-twin by prolonging pregnancy. Another example is when one twin is hydropic; ST of a hydropic twin has been reported to reverse preeclampsia when the co-twin is normal [57].

Preprocedure considerations

●Local regulations – In many states, ST is considered a form of abortion, and, as with MPR, it is important to know the law in the state/country in which the ST is performed and to be aware of any special legal considerations, such as a waiting period between signing the consent form and undergoing the procedure and gestational age constraints. Third-trimester abortion is legal in some states within the United States and in several countries.

●Fetal assessment – The first step is to correctly identify the fetus to be terminated. Previously diagnosed fetal abnormalities, particularly genetic disorders in which the fetus may appear sonographically normal, are not always obvious at the time of ST. Correct identification of the affected fetus can be made by ultrasound detection of a structural abnormality (if present); by gender (if discordant); or by previously documented fetal, gestational sac, and/or placental position. Use of dye at the time of a previously performed invasive diagnostic test is typically not useful at the time of ST since the dye may not remain in the amniotic fluid long enough. Therefore, precise in utero mapping of fetal and placental positioning at the time of prenatal diagnosis is critical; we perform a description and image documentation of the position of the sacs/placentas in three planes (axial, sagittal, and coronal).

If there is any doubt as to the correct identification of the affected fetus at the time of ST, rapid determination of fetal karyotype by fluorescent in situ hybridization (FISH) can be performed before the procedure (assuming the abnormality is detected by FISH); results of FISH are usually available within 48 hours. Although the fetuses will move around, the relationship of each sac to the other sac(s) and the position/appearance of the placenta will not change much in these 48 hours. We have found that performing FISH before ST is particularly helpful if the initial diagnostic genetic procedure was done at another institution.

●Chorioamnionicity – As with MPR, accurate determination of chorionicity is crucial before performing ST, as termination techniques are different for dichorionic versus monochorionic pairs (intrathoracic potassium chloride injection versus cord occlusion). (See 'Monochorionic fetuses' below.)

Dichorionic fetuses

Timing — We believe that in experienced hands, ST is safe and effective at any gestational age but should occur soon after the decision is made to proceed with termination. (See 'Pregnancy outcome' below.)

Technique — Potassium chloride is injected into the thoracic cavity (ideally intracardiac) of the affected fetus under ultrasound guidance until asystole is confirmed, similar to the technique used for MPR. Digoxin or lidocaine can also be used, although the time to asystole can be longer. (See 'Technique' above.)

Postprocedure follow-up — We perform an ultrasound one hour after the procedure to ensure asystole of the terminated fetus and normal cardiac activity of the remaining fetus(es). Additionally, we perform ultrasound examinations to confirm fetal well-being of the remaining fetus(es) one to two weeks after the procedure and monthly ultrasounds to monitor fetal growth in the third trimester.

Since ST is usually performed at a more advanced gestational age than MPR and with a larger fetus, the terminated fetus is generally not resorbed. Over a period of weeks to months, however, the amniotic fluid around the fetus will disappear, and the fetal tissue will become significantly compressed. The demised fetus usually delivers along with the placenta, without the patient perceiving its delivery as separate from placental expulsion. It can be sent to pathology but is usually too macerated for any meaningful evaluation. Care should be taken to make sure that there is no retained gestational material in the uterus.

Pregnancy outcome — In continuing pregnancies, outcome after ST is generally favorable, with low rates of prematurity and few maternal complications [22,58,59]. In two large series, the median gestational age at delivery after ST for continuing pregnancies (some multiple gestations) was 35.7 and 37.1 weeks [22,58].

The overall pregnancy loss rate before 24 weeks of gestation after first- or second-trimester ST was 4.0 percent in the largest single-center series of 200 patients [58] and 7.5 percent in the largest collaborative experience of 402 patients [22]. The lowest loss rate (2.4 percent) after ST is observed with a starting number of two [58]. This rate increased significantly to 11.1 percent when starting with three or more fetuses. The most striking increase in postprocedural loss rates was seen when more than one fetus was terminated: 42.9 versus 2.6 percent when only one fetus was terminated. The reason for the higher multifetal loss rate with ST than with MPR is unclear, but it may be an effect of small numbers in these series or later gestational age at the time of the procedure. (See 'Pregnancy outcomes' above.)

The effect of gestational age on postprocedure loss rate is unclear as data have been discordant. Some studies reported a trend toward higher rates of pregnancy loss and preterm delivery for ST after 20 weeks (including third-trimester ST) than for ST ≤20 weeks [22,59,60], while others observed a trend toward a higher rate of pregnancy loss with ST ≤20 weeks compared with ST performed later [58].

Monochorionic fetuses

Timing — Outcomes appear to be better if the procedure is performed after 18 weeks of gestation. Despite that, we typically perform the procedure once the decision has been made for ST. (See 'Pregnancy outcome' below.)

Technique — The conventional technique of injection of potassium chloride assumes multichorionic placentas, which do not contain interfetal vascular anastomoses. If two fetuses are monochorionic and thus have a shared circulation, and if one is reduced by injection of a toxin, then there is a chance that the toxin will also enter the co-twin's circulation and cause fetal death [61]. In addition, acute adverse hemodynamic changes could occur in the survivor, due to blood loss into the vascular system of the terminated fetus, and could result in neurologic injury. Therefore, if a fetus in a monochorionic gestation needs to be terminated, an alternative technique should be employed [62-64].

Potential approaches to ST in monochorionic twins include open surgical procedures (via hysterotomy), endoscopic procedures (fetoscopy), and ultrasound-guided needle techniques. Regardless of the approach, the most common method for ST is cord occlusion, because it decreases the risk to the co-twin since it separates the vascular systems of the fetuses.

Cord occlusion can be performed by a variety of techniques, including bipolar, laser, or photocoagulation; radiofrequency ablation; or suture ligation. Radiofrequency ablation has become the preferred cord occlusion technique [65-81]; however, the optimal technique has not been established, as most studies consist of single case reports or small case series, and technical factors (eg, cord diameter and length) may affect the choice of procedure.

Use of thinner instruments and minimizing operating time may reduce the rate of preterm birth and pregnancy loss [82]. Complication rates have decreased over time as the size of procedure needles and trocars has been reduced [83].

Postprocedure follow-up — Postprocedure follow-up is the same as for dichorionic fetuses. (See 'Postprocedure follow-up' above.)

Pregnancy outcome

●Overall outcomes – A systematic review of studies of umbilical cord occlusion for selective feticide in monochorionic twins included 345 cases performed at 14 to 35 weeks of gestation between 2000 and 2008 [84]. Major findings were:

•Survival of the co-twin was higher if the procedure was performed after 18 weeks of gestation (89 versus 69 percent).

•Survival after radiofrequency ablation, bipolar cord coagulation, laser cord coagulation, and cord ligation was 86, 82, 72, and 70 percent, respectively. The majority of co-twin deaths occurred within two weeks of the procedure. The overall rate of neonatal death was 4 percent.

•Three percent of procedures failed because of technical difficulties or complications, such as bleeding.

•Premature rupture of membranes complicated 22 percent of the pregnancies, and preterm delivery ≤34 weeks occurred in 33 percent.

•Seven percent of survivors experienced neonatal morbidity related to neurologic injury or secondary to prematurity.

●Comparison of ST procedures – A 2022 meta-analysis of 34 studies with a total of 1646 participants evaluated birth outcomes for six ST procedures in monochorionic pregnancies, including umbilical cord ligation (UCL), fetoscopic laser coagulation (FLC), interstitial laser coagulation (ILC), bipolar cord occlusion (BCO), radiofrequency ablation (RFA), and microwave ablation (MWA) [85]. The overall survival rates were similar across the six procedures (overall rate 77.1 percent, 95% CI 73.9-80.2) despite a significant difference in live birth rate by procedure. Other outcomes included:

•Fetal loss rate – BCO had the lowest fetal loss rate (11.7, 95% CI 7.4-16.8) while ILC had the highest (36.2 percent, 95% CI 24.4-49.0).  

•Live birth rate – The overall live birth rate was 82.7 percent (95% CI 78.9-86.2) with BPC having the highest (88.3 percent, 95% CI 83.2-92.6) and ILC having the lowest (63.8 percent, 95% CI 51.0-75.6).

•Neonatal death rate – The overall neonatal death rate was 1.5 percent, 95% CI 0.6-2.8 with BCO having the highest rate (5.0 percent, 95% CI 2.3-8.6, 16 studies). MWA, ILC, and UCL reporting zero deaths (one, six, and two studies, respectively).

●Specific to radiofrequency ablation –A prospective observational study of 100 consecutive cases of ST between 12.1 and 27.6 weeks of gestation using radiofrequency ablation reported the following findings [86]:

•Seventy-eight percent of patients had at least one live birth.

•The median gestational age at delivery was 35.2 weeks (range 24 to 41 weeks), and the median birth weight was 2494 g (range 694 to 4480 g).

•The spontaneous loss rate <24 weeks was 6 percent, half of which miscarried within two weeks of the procedure.

•There was no difference in survival if the procedure was performed prior to or after 20 weeks gestation.

•The co-twin demise rate was 12 percent, and, overall, there were three cases (3 percent) of cerebral injury on postprocedure magnetic resonance imaging.

Elective RFA for monochorionic diamniotic twins — Just as elective reduction for dichorionic twins remains an option to improve outcomes, elective reduction for monochorionic twins is an option at certain centers. In one single-center study, 13 monochorionic diamniotic (MCDA) twins undergoing elective MPR by RFA at a mean gestational age of 15.1±0.68 weeks were compared with 301 ongoing MCDA twins [87]. Patients who elected to undergo MPR had significantly lower rates of preterm birth at <37 weeks and a lower trend of preterm birth at <36 weeks' gestation without an increased risk of pregnancy loss. Median gestational age at delivery was significantly higher in the elective MPR group than in the ongoing pregnancy group (38 versus 35.9 weeks).

●Gestational age at delivery – The term delivery rate was higher for the elective radiofrequency ablation group compared with ongoing twin pregnancy group (gestational age, 38 weeks [interquartile range 36.1-39.1] versus 35.9 weeks [interquartile range 34.0-36.9], respectively). Those with ongoing twin gestations had an increased risk of preterm birth <37 weeks gestation but similar preterm birth rates at <34, < 32, or <28 weeks gestation.

●Pregnancy loss – All patients undergoing elective radiofrequency ablation had successful pregnancies with no pregnancy losses or terminations, while 22 patients (7.3 percent) with planned ongoing twins experienced total pregnancy loss at <24 weeks' gestation.

●Study group demographic differences – Patients who underwent elective multifetal pregnancy reduction had significantly higher maternal age and were more likely to be Asian, have undergone IVF, and had chorionic villus sampling performed. It is unclear how these factors may have impacted the study results.

Maternal complications — In a systematic review of 1239 ST procedures performed by fetoscopic radiofrequency ablation, the overall frequency of maternal complications was 5.2 percent (95% CI 3.00-7.96), primarily consisting of chorioamnionitis (19 out of 54), abruption (14 out of 54), and bleeding during the procedure (10 out of 54) [88]. In this review, preterm prelabor rupture of membranes, chorionic membrane separation, preterm labor, and preterm delivery were considered obstetric rather than maternal complications and thus were not reported.

POSTPROCEDURE LABORATORY CHANGES (MPR OR ST) — The maternal serum alpha-fetoprotein (MSAFP) concentration is typically elevated for several weeks after any fetal reduction procedure and is not necessarily indicative of fetal defects [89]. The elevation is probably caused by release of tissue or serum from the reduced fetus(es). Since MSAFP is always elevated, it should not be used to screen for neural tube defects. Instead, detailed ultrasonography should be performed. (See "Neural tube defects: Overview of prenatal screening, evaluation, and pregnancy management".)

If fetal genetic studies were not performed before MPR or ST, then maternal serum screening tests for Down syndrome, such as the quad screen, should not be performed after such procedures, as the presence of reduced fetuses will result in a high rate of abnormal results. Noninvasive prenatal diagnostic tests, which assess cell-free DNA in the maternal circulation, should also generally be avoided because DNA from the reduced fetuses can be present in the maternal circulation. Given the reduced reliability of maternal serum/blood screening tests after MPR and ST, detailed assessment of fetal development by ultrasound remains the preferred means to screen nonreduced fetuses for neural tube defects and markers associated with aneuploidy. Diagnostic testing performed on amniotic fluid or chorionic villi can also be used since neither are affected by a history of MPR or ST.

●(See "Sonographic findings associated with fetal aneuploidy".)

●(See "Neural tube defects: Overview of prenatal screening, evaluation, and pregnancy management".)

Laboratory evidence of coagulopathy has been observed, but no cases of clinically evident disseminated intravascular coagulation have been reported [90], and, in our practice, we do not routinely monitor laboratory changes after MPR or ST.

One study reported that MPR did not affect the validity of fetal fibronectin testing performed later in gestation [91].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Pregnancy termination" and "Society guideline links: Multiple gestation".)

SUMMARY AND RECOMMENDATIONS

Multifetal pregnancy reduction

●Multifetal pregnancy reduction (MPR) is a procedure in which the overall number of fetuses in the gestation is reduced by terminating one or more fetuses mostly chosen randomly, but with consideration of technical factors. (See 'Definition' above.)

●The primary justification for MPR is that the "take home" baby rate per pregnancy is increased by increasing the gestational age at birth and birth weight, thus reducing morbidity and mortality from preterm birth. The risks for some maternal complications, such as preeclampsia, are also reduced. Economic and psychological impacts of multiple gestation on families also play a role. (See 'Rationale' above.)

●MPR should be offered in triplet and higher order gestations, where a substantial reduction in early as well as late preterm birth has been demonstrated. It may be considered in twin pregnancies, where a reduction in primarily late preterm birth has been demonstrated. Reduction to a singleton may have additional advantages for women with comorbidities or a history of adverse pregnancy outcomes and may be considered for monochorionic diamniotic (MCDA) twins by radiofrequency ablation (RFA). (See 'Candidates' above.)

●MPR is usually performed between 10+0 and 13+6 weeks of gestation but can be performed later in gestation. Local laws and regulations, chorioamnionicity, and fetal screening/testing results by both invasive testing and ultrasound affect the timing and choice of procedure. Preprocedure chorionic villus sampling may be performed on the fetus(es) that is not planned for reduction and does not increase postprocedural loss rates while offering assurance of leaving behind a chromosomally normal fetus(es). Sonographic assessment of fetal growth and development prior to the procedure is also advised. (See 'Preprocedure considerations' above.)

●The fetus(es) chosen for MPR is generally closest to the anterior uterine wall and/or the fundus, as these sites are most accessible to transabdominal needle insertion. We inject potassium chloride into the thorax, which causes asystole, but other drugs can be used. (See 'Technique' above.)

●Most triplet and higher order pregnancies are reduced to twins. However, reduction to a singleton is becoming more common and is associated with good outcomes. More data are needed to determine whether reduction to a singleton significantly improves pregnancy outcome over a reduction to twins, which typically have acceptable pregnancy outcomes, particularly when compared with higher order multifetal gestations. (See 'Preprocedure considerations' above.)

●The major benefit of MPR is a reduction in preterm birth. MPR loss rates are approximately 5 percent but are impacted by operator experience, the starting numbers of fetuses (higher risk of loss at six or more fetuses), and the finishing number of fetuses (higher risk of loss with three or more fetuses). (See 'Pregnancy outcomes' above.)

●The maternal serum alpha-fetoprotein concentration is typically elevated for several weeks after any fetal reduction procedure. Laboratory evidence of coagulopathy has been observed, but no cases of clinically evident disseminated intravascular coagulation have been reported. (See 'Postprocedure laboratory changes (MPR or ST)' above.)

Selective termination

●Selective termination (ST) is a procedure in which one or more specific fetuses of a multifetal gestation are terminated due to a confirmed or suspected chromosomal, structural, or genetic abnormality that has been identified by ultrasound examination or by definitive fetal diagnostic testing. (See 'Definition' above.)

●ST avoids the live birth and long-term survival of a severely impaired child or a child with lethal abnormalities. In some cases, termination of the anomalous twin may optimize the outcome of the normal fetus. (See 'Rationale' above.)

●Loss rates after ST depend on starting and finishing numbers. (See 'Pregnancy outcome' above.)

●In monochorionic fetuses, the most common method for ST is cord occlusion, because it decreases the risk to the co-twin. (See 'Monochorionic fetuses' above.)