This page has moved. Click here to view.
The incidence of twin and higher-order multiple gestations has increased significantly over the past 15 years primarily because of the availability and increased use of ovulation-inducing drugs and newly developed assisted reproductive technologies such as in vitro fertilization. Multiple pregnancies and their various complications have, therefore, become more common (1). The incidence of twins, triplets, and higher-order multiple gestations now has reached approximately 3% of all pregnancies (2). When considering only those pregnancies resulting from assisted reproductive techniques, the rate of twin deliveries is from 25% to 30%; triplets account for 5% of deliveries, whereas the rate of higher-order multiple gestations is 0.5-1%
Multifetal gestation is associated with an increased risk of perinatal morbidity and mortality (see Tables 1-3). Women who are undergoing treatment for infertility should be aware of the risks and potential fetal and maternal complications of multiple gestation. Such women can benefit from preconceptional counseling.
Obstetrician-gynecologists should be prepared to manage, with consultation when necessary, the special problems and complications of twin and higher-order multiple gestations. Because of the many variables associated with higher-order gestations and the need for specialized care based on individual circumstances, this Educational Bulletin will focus on twin gestations.
Antepartum Management
Nutritional Considerations
It is recommended that maternal dietary intake in a multiple gestation be increased daily by approximately 300 kcal above that for a singleton pregnancy (4). Supplementation should include iron and folic acid. Although optimal weight gain for women with multiple gestations has not been determined, it has been suggested that women with twins gain 35-45 pounds (5).
Prenatal Diagnosis
The usual indications for prenatal diagnosis and counseling in a singleton pregnancy apply to twin and higher-order gestations. Because the incidence of twin gestation increases with maternal age, women with multiple gestations often are candidates for prenatal genetic diagnosis. Genetic counseling should make clear to the patient the need to obtain a sample from each fetus, the risk of a chromosomal abnormality, potential complications of the procedure, the possibility of discordant results, and the ethical and technical concerns when one fetus is found to be abnormal. Some research has shown that the combined risk of fetal chromosome abnormality is higher in dizygotic twin gestations than in a singleton gestation (6, 7). Structural anomalies are more common in monozygotic twins.
Maternal Serum Screening
Maternal serum alpha-fetoprotein (MSAFP) screening programs contribute to the detection of multiple gestations. Multiple gestation is the second most common reason after incorrect dating for the reported increase in the MSAFP level during the second trimester. Maternal serum alpha-fetoprotein screening will identify approximately 60% of unsuspected twin gestations and virtually all higher-order gestations. About 10% of pregnancies with an elevated MSAFP level will be explained by the presence of more than one fetus. The median value of MSAFP levels in twins from 14 to 20 weeks of gestation is 2.5 times that for singleton pregnancies (8). The levels in triplets and quadruplets are three and four times as high, respectively (9). Genetic screening programs, therefore, redefine an elevated MSAFP level for a twin gestation because there are two fetuses and a larger volume of placenta. Depending on the laboratory, a value greater than 4.5 multiples of the median in an uncomplicated twin gestation is abnormal, requiring further comprehensive ultrasound evaluation by an experienced ultrasonographer and possible amniocentesis for the detection of amniotic fluid alpha-fetoprotein and acetylcholinesterase.
Although maternal serum screening for neural tube defects can be useful in the twin pregnancy, its effectiveness in screening for trisomy 21 is not well defined (10). Further investigation is necessary to determine the clinical usefulness of multiple marker screening for Down syndrome in twin and higher-order pregnancies.
Amniocentesis
Although the risks may be increased, amniocentesis, using continuous ultrasound guidance, of both sacs can be performed successfully in most patients with a twin gestation (11 ). One technique is ultrasound-guided amniocentesis of each sac with approximately 1-2 mL of a dilute (approximately 0.08%) indigo carmine dye instilled into the first amniotic sac (12). A second amniocentesis of clear fluid confirms specimens from each sac. Methylene blue dye, which is associated with fetal bowel atresia and other complications, should not be used (13). If the twins are without question monozygotic, there is no reason to tap more than one sac.
Chorionic Villus Sampling
Chorionic villus sampling (CVS) is an appropriate method of first-trimester prenatal diagnosis in multiple gestations (14). The procedure is best performed by or under the supervision of an experienced operator who samples both placentas under ultrasound guidance between 10.0 and 12.9 weeks of gestation. Difficulties that can arise with CVS in twin gestations include the inability to obtain an adequate sample and contamination of one sample with tissue from the second. In approximately 1% of patients, tissue can be obtained only from one placenta. When CVS is performed at centers with experienced operators, twin-twin contamination occurs in approximately 4-6% of samples, causing prenatal diagnostic errors. When CVS or amniocentesis is performed in the twin gestation, documentation of the location of the fetuses and the membrane separating the sacs is important because discordant results can occur, and the location of the abnormal fetus may need to be known for future management.
Multifetal Reduction
The greater the number of fetuses within the uterus, the greater the risk for preterm delivery and adverse perinatal outcome. Multifetal pregnancy reduction may be performed to decrease the risk of serious perinatal morbidity and mortality associated with preterm delivery by reducing the number of fetuses. Preferably, this problem can be avoided by carefully monitoring patients receiving ovulation-inducing drugs and by minimizing the number of embryos transferred during in vitro fertilization or embryo transfer programs. The ethical dilemmas can be considerable. Patients with higher-order multiple gestations may be faced with the possibility of terminating the entire pregnancy, continuing the pregnancy and taking the risk of delivering severely preterm infants, or reducing the pregnancy in an effort to decrease the risk of perinatal morbidity and mortality.
Pregnancy loss is the main risk of multifetal pregnancy reduction and ranges from 10% to 26%. The benefit of this procedure is most clear in quadruplet and higher-order gestations because it increases the length of gestation of the surviving fetuses. It is unclear and remains to be determined whether multifetal pregnancy reduction improves long-term neonatal outcome of a triplet gestation reduced to twins. Elevations in MSAFP levels occur after selective reduction to twins; nonetheless, these patients should undergo a detailed ultrasound examination of the surviving fetuses in the middle of the second trimester.
Dizygotic twins can be discordant for congenital anomalies. Selective fetal termination allows a pregnancy to continue with a normal twin after the termination of an abnormal twin. In one study, selective fetal termination was performed successfully in 183 multifetal pregnancies, most of which were twin gestations (20). Indications for selective fetal termination were twins discordant for chromosomal abnormalities, fetal structural anomalies, or one twin affected by a Mendelian disorder. The preferred method for selective fetal termination was intracardiac injection of potassium chloride. The procedure caused loss of the entire pregnancy if there was a monochorionic placentation. The pregnancy loss rate before 24 weeks of gestation was 12.6%; an additional 3.8% of patients gave birth between 24 and 28 weeks of gestation. The authors of this study concluded that selective fetal termination for an abnormal pregnancy is a safe procedure when performed by experienced physicians provided there is a dichorionic placenta-tion. Ligation of the umbilical cord under endoscopic visualization has been performed for selective termination in a monochorionic pregnancy, but the safety and efficacy of this procedure requires further study (21).
Ultrasonography
Ultrasonography can be useful in both prenatal diagnosis and antepartum surveillance. With its use, less than 10% of twin gestations are undiagnosed before labor and delivery. Although the value of routine screening to promote early diagnosis is subject to debate, ultrasonography has a role in evaluating the progress of pregnancy once the diagnosis is established.
Screening
In the randomized clinical Routine Antenatal Diagnostic Imaging With Ultrasound (RADIUS) trial, special attention was given to multiple gestation. In the study, 129 multiple gestations were included. The RADIUS study concluded that multiple pregnancies were consistently diagnosed at an earlier gestational age in the screened group; however, this finding did not result in any overall alteration in management improving adverse perinatal outcome. This conclusion is not consistent with findings of other studies. In a 10-year study of 22,400 women having 43,000 routine ultrasound examinations involving 249 multiple gestations, earlier detection improved perinatal outcome. When multiple gestation is suspected on the basis of clinical examination, family history, a history of assisted reproduction, or an elevated MSAFP value, an ultrasound examination should be performed. Often, ultrasound examinations are performed early in gestation in twins to document viability; this examination also can be helpful in defining chorionicity.
Evaluation
A detailed ultrasound evaluation of a multiple gestation should be performed during the second trimester. This examination should include determination of placentation, amnionicity, and chorionicity; the number of fetuses; evaluation for fetal amniotic fluid or placental abnormalities; and an assessment of the growth of each fetus. If two separate placentas are identified or if the fetuses are of different sex, the placentation is dichorionic. A thick membrane also suggests dichorionicity. When a thin, wispy membrane is seen between two sacs with a single placenta, and the fetuses are of the same sex, monochorionicity is suggested. "Stuck twin syndrome," where there is polyhydramnios for one twin and extreme oligohydramnios for the other, can be very difficult to differentiate from a monochorionic pregnancy because the membrane may be so near the stuck twin that it is difficult to detect. Determination of chorionicity is most accurate in the first trimester; as the pregnancy progresses, it becomes less accurate.
Visualization of a membrane confirms the diagnosis of a diamniotic gestation. When dichorionicity is diagnosed, the pregnancy must also be diamniotic. The number of fetuses should be identified; this number, however, may decrease as the pregnancy progresses. Until the third trimester, twins follow the same growth curves that apply to singleton pregnancies. There is no clear clinical advantage to the use of twin- or triplet-specific ultrasound growth tables. Evaluation of serial fetal growth should include estimated fetal weight of each fetus and appropriate and concordant interval growth.
Routine Cervical Evaluation
Routine cervical evaluation by either clinical or ultra-sonographic assessment has been investigated as an approach to predict preterm birth in the multiple gestation. A cervical scoring system, calculated as cervical length (in centimeters) minus cervical dilatation at the internal os (in centimeters) has been used. This system has been associated with a positive predictive value of 75%, which is associated with a fourfold increased relative risk of preterm delivery. Weekly digital cervical examination for clinical assessment of the cervix has not been associated with adverse maternal or fetal outcome.
Vaginal ultrasonography has been used to measure cervical width, length, and funneling and to examine the relationship of these measurements to the risk of preterm birth. In this study, the risk of spontaneous preterm delivery was increased in women who were found to have a short cervix. Although this is a promising technique, further evaluation of transvaginal ultrasonography by a prospective randomized study is necessary to determine its role in the prevention of preterm birth.
Several investigators have studied the use of cervical cerclage in multiple gestations to prevent preterm birth. Because there is no clear-cut benefit to this approach, and its use is associated with both maternal and fetal risks, this procedure is not recommended for multiple gestations.
Bed Rest
The role and value of bed rest at home or in the hospital in the prevention of preterm delivery of a multiple gestation remains controversial. Not only is hospital bed rest costly, stressful, and disruptive, there is no clear consensus that it is of any benefit. Numerous studies have failed to show that bed rest decreases the incidence of preterm delivery, lengthens gestation, or improves neonatal morbidity in multiple gestation. Also, there are little data to support activity reduction in multiple or singleton gestations. Antepartum hospitalization may be necessary in the multiple gestation for managing complications such as preterm labor and abnormal fetal growth.
Antepartum Surveillance
The routine use and benefit of antepartum fetal surveillance in the uncomplicated multifetal gestation has not been shown to be of benefit. When intrauterine growth restriction, abnormal fluid volumes, growth discordance, pregnancy-induced hypertension, fetal anomalies, monoamnionicity, or other pregnancy complications occur, fetal surveillance, including nonstress testing or the modified or standard biophysical profile, is indicated. The biophysical profile is as reliable in multiple gestations as in singleton gestations.
Although some patients may find it difficult to distinguish between fetal movement of each twin, fetal movement counting can be an adjunct to these antepartum surveillance techniques. Umbilical cord velocimetry may be helpful in evaluating the severely growth-restricted fetus, but its role in antepartum fetal surveillance of the singleton or multiple gestation is yet to be determined.
Home Uterine Activity Monitoring
The use of the home uterine activity monitor has been advocated by some to prevent or manage preterm labor in singleton or multiple gestations. There are, however, no data to support its use.
Management of Complications
Preterm Labor
The most significant and common complication of multiple pregnancy is preterm labor resulting in preterm delivery. Perinatal morbidity and mortality arc affected by gestational age and weight at delivery as well as by the number of fetuses. The presence of a single anomalous fetus in a twin gestation increases the risk of preterm delivery compared with nonanomalous twin gestations. Patient education, risk assessment, serial cervical evaluation by manual or transvaginal ultrasound examinations, specialized antepartum
Preterm Rupture of Membranes
Preterm premature rupture of membranes occurs more
frequently in twin gestations than in singleton gestations. A matched-control study of preterm premature rupture of membranes in twins concluded that the nonpresenting twin more frequently had hyaline membrane disease, respiratory complications, and required more oxygen therapy than the presenting infant. There were no significant differences between twin and singleton gestations in infectious morbidity, cord prolapse, abruptio placentae, or latency to delivery. Infant morbidity and mortality were high for the nonpresenting infant at significantly increased risk of respiratory complications.
Preterm rupture of the presenting sac occurs most frequently. Although the incidence is unknown, membrane rupture of the nonpresenting sac also can occur. When preterm rupture of membranes occurs in the multiple gestation, it most frequently occurs prior to fetal lung maturation. Preterm premature rupture of membranes occurs more frequently in the triplet gestation than in the twin
Fetal Disorders
Intrauterine Growth Restriction
Intrauterine growth restriction occurs more frequently in twins and higher-order multiple gestations than in a singleton pregnancy and is a significant cause of increased neonatal morbidity and mortality. Antepartum management of the multiple gestation should include early identification of intrauterine growth restriction by ultrasonography and increased fetal surveillance to improve perinatal outcome. There is no single definition for the diagnosis of intrauterine growth restriction or discordant growth in twins. Estimated fetal weight is helpful in making the diagnosis of intrauterine growth restriction and discordance between fetuses.
Intrauterine growth restriction usually is diagnosed either when the estimated fetal weight decreases below the 10th percentile for a singleton gestation or when there is discordance (ie, a difference in estimated fetal weight of greater than 20% between twin A and twin B expressed as a percentage of the larger twin' s weight). Once intrauterine growth restriction has been diagnosed, the multiple
Other Complications
Twin-twin transfusion syndrome, monoamniotic twinning, conjoined twins, and acardia (or twin reversed arterial per-fusion sequence) are fetal complications of monochorionic gestations that are rarely encountered by the obstetrician-gynecologist. Patients with these complications should be cared for by or in collaboration with specialists familiar with
Death of One Twin
Death of one fetus of a multiple gestation can occur at any time during pregnancy although it is more common in the first trimester. These patients can benefit from perinatal grief counseling. Only 50% of twin gestations diagnosed in the first trimester result in delivery of two live infants. Early fetal demise of one twin may not be recognized clinically but may be diagnosed by first-trimester ultrasonography. A documented multiple gestation that spontaneously loses one or more fetuses during the first trimester is called the
Intrapartum Management
Timing
Ideally, women with multiple pregnancies should undergo delivery by 40 weeks of gestation; however, the clinician must frequently weigh the risks to the fetuses if intrauterine life is continued versus the risk to the mother if the pregnancy is continued. For example, preterm labor, discordance, abnormal fetal surveillance, or pregnancy-induced hypertension may mandate earlier intervention. The ideal time of delivery for uncomplicated pregnancies is uncertain; however, if elective delivery is considered before 38 weeks of gestation, fetal lung maturity should be assessed. Although there are few data pertaining to multiple gestations beyond 40 weeks of gestation, delivery should probably be effected by this time.
Labor and Delivery
When a woman with a known or suspected multiple gestation presents in labor, confirmation as soon as possible by ultrasound examination of fetal number and presentations is indicated. Both twins should be monitored continuously during labor. Ultrasonography should be available to determine the heart rate of the second twin as well as its orientation following delivery of the first twin. Appropriate, experienced pediatric and anesthesia personnel should be notified and available at delivery. Capability for emergency cesarean delivery is necessary and blood should be available because the likelihood of operative intervention, as well as postpartum hemorrhage, is increased.
Route of Delivery
Controversy surrounds the preferred route of delivery for some multiple gestations, especially twins. Delivery should be based on individual needs and may depend on the clinician's practice and experience. The various twin presentations should be taken into account. All combinations of twin presentations and their frequency essentially can be classified into the following groups:
Twin A-Vertex with Twin B-Vertex. Vaginal delivery is anticipated for vertex-vertex twins. Cesarean delivery should only be performed for the same indications applied to singleton gestations. In one series, 81.2% of the vertex-vertex twin gestations were successfully delivered vaginally.
Twin A-Vertex with Twin B-Nonvertex. There are conflicting data on the management of twins in vertex-breech or vertex-transverse presentation. Because depressed Apgar scores and an increased perinatal mortality rate have been associated with vaginal delivery of the second twin, cesarean delivery has been advocated by some whenever the second twin is in a nonvertex presentation. Cesarean delivery, however, is not always necessary. Vaginal delivery of twin B in the nonvertex presentation is a reasonable option for a neonate with an estimated weight greater than 1,500 g. Under these circumstances, it has been reported that
Interval Between Deliveries
The interval between delivery of each twin is not critical in determining the outcome of twin B. Surveillance of twin B with real-time ultrasonography and continuous monitoring of the fetal heart rate, however, are advised after the delivery of twin A. Rapid delivery may be required because of complications, such as abruptio placentae, cord prolapse, or a decrease in the fetal heart rate. If labor has not resumed within a reasonable time after the delivery of twin A, oxytocin augmentation with careful fetal heart rate surveillance