Obstetrics & Gynecology:
Perinatal Outcome of Monoamniotic Twin Pregnancies
Hack, Karien E. MD, PhD1; Derks, Jan B. MD, PhD1; Schaap, Arty H. MD, PhD3; Lopriore, Enrico MD, PhD4; Elias, Sjoerd G. MD, PhD6; Arabin, Birgit MD, PhD7; Eggink, Alex J. MD, PhD8; Sollie, Krystyna M. MD, PhD9; Mol, Ben Willem J. MD, PhD10; Duvekot, Hans J. MD, PhD11; Willekes, Christine MD, PhD12; Go, Attie T. MD, PhD13; Koopman-Esseboom, Corine MD, PhD2; Vandenbussche, Frank P. MD, PhD5; Visser, Gerard H. MD, PhD1
From the 1Department of Obstetrics, 2Department of Neonatology, Wilhelmina Children's Hospital, University Medical Centre Utrecht, Utrecht, 3Department of Obstetrics, Academic Medical Center, Amsterdam, 4Department of Neonatology, 5Department of Obstetrics, Leiden University Medical Center, Leiden, 6Julius Centre for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht, 7Department of Obstetrics, Isala Clinics, Zwolle, 8Department of Obstetrics, Radboud University Nijmegen Medical Center, Nijmegen, 9Department of Obstetrics, University Medical Center Groningen, Groningen, 10Department of Obstetrics, Maxima Medical Center, Veldhoven, 11Department of Obstetrics, Erasmus Medical Center, Rotterdam, 12Department of Obstetrics, University Hospital Maastricht, Maastricht, and 13Department of Obstetrics, Vrije Universiteit Medical Center, Amsterdam, the Netherlands.
Corresponding author: Dr. K. E. A. Hack, Department of Obstetrics, Wilhelmina Children's Hospital, University Medical Centre Utrecht, KE.04.123.1, PO Box 85090, 3508 AB Utrecht, the Netherlands; e-mail: email@example.com.
Financial Disclosure The authors did not report any potential conflicts of interest.
OBJECTIVE: To study perinatal mortality and neonatal morbidity in a large cohort of monoamniotic twin pregnancies with special emphasis to the gestational age-specific mortality.
METHODS: The study included monoamniotic twin pregnancies delivered in 10 perinatal centers in the Netherlands between January 2000 and December 2007.
RESULTS: A total of 98 monoamniotic pregnancies were included. The perinatal mortality rate (20 weeks of gestation through 28 days of life) was 19%; after exclusion of fetuses with lethal anomalies, the rate was 17%. After 32 weeks of gestation, only two pregnancies were complicated by perinatal mortality (4%). The incidence of twin–twin transfusion syndrome was 6%. The incidence of congenital heart anomalies and cerebral injury was 4% and 5%, respectively.
CONCLUSION: The current incidence of perinatal mortality in monoamniotic twins is considerably lower than in previous decades, but it is still high and occurs throughout pregnancy.
LEVEL OF EVIDENCE: III
Monoamniotic twinning is a rare event and occurs in approximately 1% of all monozygotic twin gestations. Monoamniotic twins are associated with high antenatal and perinatal mortality rates. In the past, mortality rates ranging from 30% to 70% have been reported.1–4 However, more recent publications suggest a substantially improved perinatal survival with mortality rates of 10% to 20%.5 Although the high mortality rate is partly attributable to common complications in (monochorionic) twin pregnancies, such as preterm delivery and low birth weight, the most important cause of death is entanglement and knotting of the umbilical cords, a specific complication in monoamniotic twins. Cord entanglement has been reported in up to 70% of monoamniotic twins with 50% or more of deaths attributed to this complication.6 Due to the rare occurrence of monoamniotic twinning, only relatively small case series have been published with varying survival data, including diverse mortality rates after 30–32 weeks of gestation and varying recommendation on the need to perform early delivery.7–11
In a large series of 98 monoamniotic twin pregnancies, we evaluated perinatal mortality and morbidity with special emphasis to the gestational-age specific mortality.
MATERIALS AND METHODS
The medical records of all monoamniotic twin pregnancies delivered at 10 perinatal centers in the Netherlands between January 2000 and December 2007 were identified and reviewed. The study was approved by the institutional review board of the University Medical Center Utrecht, and local permission was obtained in the other centers. Monoamnionicity was determined on the basis of first-trimester ultrasound features of the absence of a dividing amniotic membrane with a single placenta and concordant gender and/or confirmed by the attending obstetrician with postpartum examination of placentas and intertwin membranes and histopathological examination by a specialized pathologist. Acardiac twins were excluded from analysis.
Gestational age was calculated from the first day of the last menstrual period and confirmed by first-trimester ultrasonography. Due to the rarity of these pregnancies, local antenatal management and surveillance of the monoamniotic twin pregnancies differed between the study centers. In general, all monoamniotic twin gestations were monitored by regular ultrasound assessment of growth and amniotic fluid volume, Doppler of the umbilical artery, and cord entanglement fortnightly. Subjects with either nonreassuring fetal findings (fetal heart rate abnormalities, abnormal Doppler findings, suspected intrauterine growth restriction-defined as an estimated fetal weight below the 10th centile) or with maternal complications were submitted to frequent but at least twice weekly maternal–fetal evaluations that were performed during hospitalization or during visits at an outpatient clinic setting. In four centers, women were hospitalized between 30 and 32 weeks of gestation to monitor fetal heart rates twice a day. In most centers, elective cesarean delivery was offered around 32–34 weeks of gestation after steroid treatment or determination of lung maturity.
The diagnosis of twin–twin transfusion syndrome was made by prenatal ultrasound criteria.12 Since the hydramnios-oligohydramnios sequence cannot be detected in monoamniotic pregnancies, diagnosis of twin–twin transfusion syndrome was based on the identification of other clinical manifestations of the syndrome, such as hydramnios (deepest vertical pocket 8 or more cm before 20 weeks of gestation or 10 or more cm after 20 weeks of gestation), discordance in bladder size, and abnormal Doppler flow patterns in either twin.
Stillbirth was defined as intrauterine death of a fetus 20 or more weeks of gestation. Gestational age at time of stillbirth was ascertained by ultrasonography. Early neonatal death was defined as death of a neonate during the first 7 days of life, whereas late neonatal death was defined as death between 8 and 28 days after birth. Overall perinatal mortality was defined as stillbirth or neonatal death (28 or fewer days after birth). We recorded the presence of congenital heart malformations and severe cerebral injury on cranial ultrasonography. Major neuromorbidity was defined as the occurrence of one of the following abnormalities on cerebral ultrasonography: cystic periventricular leukomalacia,13 intraventricular hemorrhage greater than grade II,14 cerebral artery infarction, posthypoxic ischemic encephalopathy, ventriculomegaly, or congenital hydrocephalus.
The main objective of the analysis was to study perinatal mortality. We constructed Kaplan Meier curves, in which we assessed both time to delivery and time to fetal death. Mortality rates were calculated per week of gestation; we constructed a table in which, for each gestational week, perinatal mortality was expressed in relation to the number of women at risk at the beginning of the week. All statistical analyses were performed with the SPSS 12.0 statistical package (SPSS Inc., Chicago, IL). Descriptive statistics were used to explore data and characterize the study population. Kaplan-Meier analysis was used to estimate cumulative survival and confidence intervals (CIs) for estimated proportions were calculated.
We identified 103 monoamniotic pregnancies from which five acardiac twins were excluded. Table 1 shows the baseline characteristics of the study population. Nine pregnancies were thought to be monochorionic diamniotic and only recognized as monoamniotic at delivery. The monoamnionicity in the other 89 pregnancies was identified antenatally. The proportion of female twin pairs was higher (65%) than the proportion of male twin pairs (35%). Six pregnancies were complicated by twin–twin transfusion syndrome (6%). There was no increase in incidence of monoamniotic twins in the course of study period.
Six pregnancies were complicated by death of both fetuses before 20 weeks of gestation. In two pregnancies, mortality was caused by cord entanglement and strangulation; in the other four pregnancies, the cause of death was unknown. Details on the fetal deaths occurring 20 or more weeks of gestation, including possible cause, are presented in Table 2. There were 22 intrauterine deaths, of which eight double fetal deaths and six single fetal deaths. Three of the single fetal deaths were followed by neonatal death of the second twin (all three due to cerebral artery infarction). No cerebral damage was found in the remaining cotwin death survivors (n=3).
In the neonatal period, there were 12 neonatal deaths, 10 single neonatal deaths, and one double neonatal death (Table 3). Seven deaths were caused by lethal congenital malformations. The overall perinatal mortality rate (20 weeks of gestation through 28 days of life) was 19% (95% CI 12–24%). After exclusion of lethal anomalies, overall perinatal mortality was 17% (95% CI 11–23%). Table 4 lists the mode of delivery and the indications for delivery. Forty percent of neonates were born vaginally, half of which were either previable or were part of a nonrecognized monoamniotic twin pregnancy. Only two neonates were born by cesarean delivery after a trial of delivery, both being the second of a twin, and in both cases this was due to heart rate decelerations and cord entanglement. There was one neonatal death in the vaginal delivery group due to meningitis/sepsis (5.3%), compared with four neonatal death in the group born by planned cesarean delivery (7.0%).
Figure 1 shows the survival curve of nonanomalous prenatally recognized monoamniotic twins.
Table 5 shows the perinatal outcome of the twins. Eighty-seven of 164 liveborn neonates (53%, 95% CI 45–61%) were admitted to the neonatal intensive care unit for a median length of 6 days (range 1–91 days). Seven neonates had a congenital heart anomaly (4%, 95% CI 1–7%); two had a coarctation of the aorta, two had a ventricular septal defect, two infants had a transposition of the great arteries (of which one in combination with a ventricular septal defect), and one neonate had tricuspid valve insufficiency. The incidence of severe neonatal cerebral abnormalities was 5% (8 of 164, 95% CI 2–8%; cranial ultrasound data available for all but six neonates, all born at term with no signs of neuromorbidity).
Nine pregnancies were thought to be monochorionic diamniotic and only recognized as monoamniotic at delivery. Median gestational age at delivery was 36 1/7 weeks (range 26 4/7 to 38 0/7 weeks). Four of these pregnancies were delivered before 34 weeks of gestation, all with spontaneous onset of preterm labor. Five pregnancies were delivered after 34 weeks of gestation; three women had an induction of labor due to monochorionicity, and two delivered after spontaneous onset of labor. One neonate died (born at 27 5/7 weeks) due to necrotizing enterocolitis. Monoamniotic twin pregnancies that turned out to be monochorionic diamniotic (either later in pregnancy or at time of delivery) were not recorded.
Data were obtained from the 10 level-3 perinatal centers and include approximately 40% of all monoamniotic twins delivered in the Netherlands in that time period. It is therefore not a population-based study and outcome in the other 60% of cases may have been different.
There was a high incidence of perinatal mortality and neonatal morbidity in nonanomalous monoamniotic infants. The incidence of twin–twin transfusion syndrome in this cohort was 6%. Congenital heart anomalies and severe cerebral injury occurred in 4% and 5% of monoamniotic twin infants, respectively.
In the past, perinatal mortality rate was reported to range between 30% and 70%.1–4 More recent studies (case series and literature reviews) suggest a substantially improved perinatal survival with reported mortality rates of 10% to 20%,7–11 which is in agreement with our findings. The reported incidences of perinatal mortality vary widely due to differences in ascertainment, prenatal recognition, inclusion of fetal anomalies, and individual unit practice. All published data are relatively small observational studies and open to a publication bias (both negative as positive). The presence of fetal abnormalities contributes significantly to adverse outcome data, and it is important to distinguish outcomes between these subgroups. In our study, perinatal mortality and corrected perinatal mortality rates were 19% and 17%, respectively. This was slightly higher than the nonanomalous perinatal mortality rate reported by Heybourne et al15 (12.6%). Compared with the older literature, when intensive antepartum management and early delivery were not performed in these high-risk pregnancies, perinatal mortality has decreased significantly. We assume that intensive monitoring (ie, more frequently) and earlier delivery has contributed to this decrease in mortality. However, these conclusions cannot directly be made from the data presented, mainly due to the retrospective character of our study and the heterogeneity with respect to antepartum management. The lack of standardization of antepartum management and the timing of delivery is a limitation of this study.
Entanglement and knots of the umbilical cords are the major cause of fetal death in monoamniotic twins.8 It may be detected by prenatal ultrasonography already in the first trimester.8,18,19 It is hypothesized that initiation of cord entanglement is a phenomenon of early pregnancy, when amniotic fluid volume in relation to the fetal mass is greater,20 and that location of the cord entanglement may influence the risk of antepartum death. (Arabin B, HackKEA Does the location of cord entanglement matter for ante partum death in monoamniotic twins? Ultrasound Obstet Gynecol. In press).
Congenital malformations are found in 15% to 20% of monozygotic twin pregnancies.21 Acardiac twinning, anencephaly, and congenital heart defects are typically related to monoamniotic twins. A proportion of these defects are acquired as a result of the altered hemodynamics in the recipient twin associated with twin–twin transfusion syndrome. Karatza et al22 reported a 2% incidence of congenital cardiac abnormalities in 89 monochorionic twin pregnancies unaffected by twin–twin transfusion syndrome. In another study of 165 monochorionic twin pregnancies,23 the overall risk of at least one fetus in a monochorionic twin pair having a structural congenital cardiac anomaly was 9%. In a small subgroup of 7 monoamniotic twins, five children had a congenital heart malformation (36%). Ventricular septal defect was the most common lesion diagnosed. In our study, the incidence of congenital heart malformations was 4%. Since the incidence of heart anomalies and cerebral ultrasound abnormalities is high among monoamniotic twins, all monoamniotic infants (especially after death of the cotwin) should be examined postnatally, including a cerebral and cardiac ultrasound scan. In our study, three of the single fetal deaths were followed by neonatal death of the second twin due to cerebral artery infarction. This knowledge could alter counseling of parents with single fetal death with regard to the possibility of severe neuromorbidity, either leading to neonatal death or (severe) handicap of the second twin.
The incidence of twin–twin transfusion syndrome in this cohort of monoamniotic twins was low (6%), which is in agreement with other reports (incidence ranging from 3% to 10%15, 24–26). In contrast, the incidence of twin–twin transfusion syndrome in monochorionic diamniotic twins is higher (10% to 15%).27,28 The difference in the rate of twin–twin transfusion syndrome is partly due to a different anastomotic pattern between monoamniotic placentas and monochorionic diamniotic placentas. Almost all monoamniotic placentas have arterio-arterial anastomoses.29 The presence of these arterio-arterial anastomoses protects against hemodynamic disequilibrium by allowing intertwin blood flow and hence the development of twin–twin transfusion syndrome.30 Moreover, in monoamniotic pregnancies, twin–twin transfusion syndrome cannot be diagnosed using the standard criteria, since it is impossible to diagnose the occurrence of the hydramnios-oligohydramnios sequence in monoamniotic pregnancies. Delayed or lack of identification of clinical manifestations of twin–twin transfusion syndrome may also account for the reduced rate of twin–twin transfusion syndrome in monoamniotic twins.31 There is an urgent need for clear diagnostic criteria of twin–twin transfusion syndrome in monoamniotic twins. More studies are needed to determine standardized criteria.
Recommended timing of delivery ranges between 32 and 35 weeks of gestation,7,8,11 after lung maturation has been demonstrated by amniocentesis or after a course of corticosteroids to enhance pulmonary maturation. Perinatal deaths may occur throughout pregnancy, and fetal mortality cannot always been foreseen. In contrast to two studies in which no mortality after 32 weeks of gestation was found,9,10 we report four deaths in continuing pregnancies after 32 weeks (one double intrauterine death and an intrauterine death with neonatal death of the cotwin due to severe cerebral damage caused by hemodynamic problems after intrauterine death of the sibling; all intrauterine deaths were caused by entanglement and true knots of the umbilical cords). This is in agreement with most publications on this topic.11,30,31 Delivery of all monoamniotic twins at 32 weeks of gestation might have saved four infants (4% of ongoing pregnancies at that age). However, this should be weighed against the risks of neonatal respiratory disorders. It should also be weighed against a misdiagnosis of a twin considered to be monoamniotic , but in fact monochorionic diamniotic, since in the latter group the risk of intrauterine death is lower, not warranting delivery at such an early age.32–35 Due to the retrospective nature of our study, we were unable to assess the incidence of such a misdiagnosis.
In our series, 40% of infants were born vaginally, since it was not routine in all hospitals to deliver monoamniotic twins by cesarean. The risks of a vaginal delivery relate primarily to cord entanglement and cord compression during labor and especially after delivery of the first infant. Therefore most authors favor cesarean delivery in monoamniotic twins, but vaginal delivery has been reported in several case reports.10,36 In our series, in two cases, a cesarean delivery of the second neonate was done due to problems after the vaginal birth of the first neonate, but none of the neonates died due to labor related complications. However, our data set was relatively small and does not allow definitive answers. At present, we would favor a cesarean delivery in all monoamniotic twins. In summary, the incidence of perinatal mortality in monoamniotic twins has decreased over the years, but remains high (15% to 20%) and occurs throughout pregnancy. There is no consensus about optimal antenatal management and timing of delivery. We assume that intensive monitoring (ie, more frequently) and earlier delivery has contributed to this decrease in perinatal mortality.
1. Raphael SI. Monoamniotic twin pregnancy: a review of the literature and a report of 5 new cases. Am J Obstet Gynecol 1961;81:323–30.
2. Wensinger JA, Daly RF. Monoamniotic twins. Am J Obstet Gynecol 1962;83:1254–6.
3. Timmons JD, Dealvarez RR. Monoamniotic twin pregnancy. Am J Obstet Gynecol 1963;86:875–81.
4. Benirschke K. The placenta in twin gestation. Clin Obstet Gynecol 1990;33:18–31.
5. Dickinson JE. Monoamniotic twin pregnancy: a review of contemporary practice. Aust N Z J Obstet Gynaecol 2005;45:474–8.
6. Su LL. Monoamniotic twins: diagnosis and management. Acta Obstet Gynecol Scand 2002;81:995–1000.
7. Rodis JF, McIlveen PF, Egan JF, Borgida AF, Turner GW, Campbell WA. Monoamniotic twins: improved perinatal survival with accurate prenatal diagnosis and antenatal fetal surveillance. Am J Obstet Gynecol 1997;177:1046–9.
8. Aisenbrey GA, Catanzarite VA, Hurley TJ, Spiegel JH, Schrimmer DB, Mendoza A. Monoamniotic and pseudomonoamniotic twins: sonographic diagnosis, detection of cord entanglement, and obstetric management. Obstet Gynecol 1995;86:218–22.
9. Carr SR, Aronson MP, Coustan DR. Survival rates of monoamniotic twins do not decrease after 30 weeks' gestation. Am J Obstet Gynecol 1990;163:719–22.
10. Tessen JA, Zlatnik FJ. Monoamniotic twins: a retrospective controlled study. Obstet Gynecol 1991;77:832–4.
11. Roque H, Gillen-Goldstein J, Funai E, Young BK, Lockwood CJ. Perinatal outcomes in monoamniotic gestations. J Matern Fetal Neonatal Med 2003;13:414–21.
12. Huber A, Hecher K. How can we diagnose and manage twin-twin transfusion syndrome? Best Pract Res Clin Obstet Gynaecol 2004;18:543–56.
13. de Vries LS, Eken P, Dubowitz LM. The spectrum of leukomalacia using cranial ultrasound. Behav Brain Res 1992;49:1–6.
14. Papile LA, Burstein J, Burstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. J Pediatr 1978;92:529–34.
15. Heyborne KD, Porreco RP, Garite TJ, Phair K, Abril D. Improved perinatal survival of monoamniotic twins with intensive inpatient monitoring. Am J Obstet Gynecol 2005;192:96–101.
16. Allen VM, Windrim R, Barrett J, Ohlsson A. Management of monoamniotic twin pregnancies: a case series and systematic review of the literature. BJOG 2001;108:931–6.
17. Beasley E, Megerian G, Gerson A, Roberts NS. Monoamniotic twins: case series and proposal for antenatal management. Obstet Gynecol 1999;93:130–4.
18. Arabin B, Laurini RN, van Eyck J. Early prenatal diagnosis of cord entanglement in monoamniotic multiple pregnancies. Ultrasound Obstet Gynecol 1999;13:181–6.
19. Sebire NJ, Souka A, Skentou H, Geerts L, Nicolaides KH. First trimester diagnosis of monoamniotic twin pregnancies. Ultrasound Obstet Gynecol 2000;16:223–5.
20. Overton TG, Denbow ML, Duncan KR, Fisk NM. First-trimester cord entanglement in monoamniotic twins. Ultrasound Obstet Gynecol 1999;13:140–2.
21. Baldwin V. The pathology of monochorionic monozygosity. In: Baldwin V, editor. Pathology of multiple pregnancy. New York (NY):Springer; 1994. p.199–214.
22. Karatza AA, Wolfenden JL, Taylor MJ, Wee L, Fisk NM, Gardiner HM. Influence of twin-twin transfusion syndrome on fetal cardiovascular structure and function: prospective case-control study of 136 monochorionic twin pregnancies. Heart 2002;88:271–7.
23. Manning N, Archer N. A study to determine the incidence of structural congenital heart disease in monochorionic twins. Prenat Diagn 2006;26:1062–4.
24. Ezra Y, Shveiky D, Ophir E, Nadjari M, Eisenberg VH, Samueloff A, et al. Intensive management and early delivery reduce antenatal mortality in monoamniotic twin pregnancies. Acta Obstet Gynecol Scand 2005;84:432–5.
25. Demaria F, Goffinet F, Kayem G, Tsatsaris V, Hessabi M, Cabrol D. Monoamniotic twin pregnancies: antenatal management and perinatal results of 19 consecutive cases. BJOG 2004;111:22–6.
26. Suzuki S, Kaneko K, Shin S, Araki T. Incidence of intrauterine complications in monoamniotic twin gestation. Arch Gynecol Obstet 2001;265:57–9.
27. Lutfi S, Allen VM, Fahey J, O'Connell CM, Vincer MJ. Twin-twin transfusion syndrome: a population-based study. Obstet Gynecol 2004;104:1289–97.
28. Lewi L, Jani J, Blickstein I, Huber A, Gucciardo L, Van Mieghem T, et al The outcome of monochorionic diamniotic twin gestations in the era of invasive fetal therapy: a prospective cohort study. Am J Obstet Gynecol 2008;199:514.e1–8.
29. Umur A, van Gemert MJ, Nikkels PG. Monoamniotic-versus diamniotic-monochorionic twin placentas: anastomoses and twin-twin transfusion syndrome. Am J Obstet Gynecol 2003;189:1325–9.
30. Denbow ML, Cox P, Taylor M, Hammal DM, Fisk NM. Placental angioarchitecture in monochorionic twin pregnancies: relationship to fetal growth, fetofetal transfusion syndrome, and pregnancy outcome. Am J Obstet Gynecol 2000;182:417–26.
31. van den Wijngaard JP, Umur A, Ross MG, van Gemert MJ. Modelling the influence of amnionicity on the severity of twin-twin transfusion syndrome in monochorionic twin pregnancies. Phys Med Biol 2004;49:N57–64.
32. Hack KEA, Derks JB, Elias SG, Franx A, Roos EJ, Voerman SK, et al. Increased perinatal mortality and morbidity in monochorionic versus dichorionic twin pregnancies: clinical implications of a large Dutch cohort study. BJOG 2008;115:58–67.
33. Simoes T, Amaral N, Lerman R, Ribeiro F, Dias E, Blickstein I. Prospective risk of intrauterine death of monochorionic-diamniotic twins. Am J Obstet Gynecol 2006;195:134–9.
34. Lewi L, Jani J, Blickstein I, Huber A, Gucciardo L, Van Mieghem T, et al. The outcome of monochorionic diamniotic twin gestations in the era of invasive fetal surgery: a prospective cohort. Am J Obstet Gynecol 2008;199:514.e1–8.
35. Acosta-Rojas R, Becker J, Munoz-Abellana B, Ruiz C, Carreras E, Gratacos E. Twin chorionicity and the risk of adverse perinatal outcome. Int J Gynaecol Obstet 2007;96:98–102.
36. Dubecq F, Dufour PH, Vinatier D, Thibault D, Lefebvre C, Tordjeman N, et al. Monoamniotic twin pregnancies: review of literature, and a case report with vaginal delivery. Eur J Obstet Gynecol Reprod Biol 1996;66:183–6.
Figure. No caption available.
© 2009 by The American College of Obstetricians and Gynecologists.