Opinion - Monochorionic twin pregnancies: 'les liaisons dangereuses'  Dr. Yves Ville,  University Hospital of Poissy, France Ultrasound obstet. Gynecol. 10 (1997) 82-85  If Rebecca had had an ultrasound examination performed in the first trimester of her pregnancy when she was expecting Esau and Jacob, the confusion and controversy about the type of twinning and their complications might not have invaded the medical literature. Indeed, they have often been described as dizygotic because of their physical dissimilarity as adults; but on the other hand, because Esau came out red, it has also been reported that he was the first example of the feto-fetal transfusion syndrome (TTS). Although not quoted in Genesis, but in this issue of the Journal (2), Sebire and colleagues have managed to put a square plug in a round hole by bringing together their main research topics for the last 6 years: nuchal translucency thickness measurement and severe twin-twin transfusion syndrome. In an impressive series of 132 monochorionic twin pregnancies, including 16 that developed severe TTS before 24 weeks' gestation, the authors have assessed the positive and negative predictive values of nuchal translucency measurement on the occurrence of TTS in the second trimester of pregnancy. These were 380 and 91%, respectively, with an incidence of TTS of 12%. The idea developed by Sebire and colleagues of the increased nuchal translucency being the expression of an early hemodynamic imbalance is appealing. The authors base their interpretation on the association they have previously described between increased nuchal translucency and congenital cardiac malformations which they interpreted as the early hemodynamic expression of the cardiac defect in utero. The association of increased nuchal translucency and later development of TTS is therefore presented in this paper as the expression of a hemodynamic imbalance without cardiac defect. The hypothesis is elegant, but the short Cut taken with this argument would need to be strengthened. Although the sonographic expression of hypo- or hyper volemia is probably not unequivocal in the first trimester, it would have been interesting to identify which twin was going to become the recipient and which twin the donor, respectively. Their account does not state whether the hearts of these twins were assessed by a perinatal cardiologist. Indeed, although a twice higher incidence of congenital heart defects reported in monozygotic twins, ascompared to dizygotic twins, could be biased in many ways anatomical and functional cardiac abnormalities have been described specifically in recipient twins in TTS, involving mainly pulmonary stenosis (4'5). Considering the hemodynamic status of the recipient twin, the most likely hypothesis is an increased right ventricular afterload with secondary functional pulmonary stenosis due to hypertrophic right cardiomyopathy. However, Sebire and colleagues' report might support the hypothesis of a primary cardiac defect that could be transient (6), remain asymptomarie or progress towards a hypoplastic right heart. The incidence of these possible developments cannot be drawn from the literature, but the increasing number of observations presented at perinatology meetings should be collected. To what extent this could be suggested as a primary event preceding hemodynamic imbalance in TTS cannot be assessed, but it suggests that serial cardiac evaluation of monochorionic twins in utero could provide information to answer this question. Indeed, the pathophysiology of TTS is poorly understood, and, although transfusion has been confirmed in vivo (7), the pathophysiology of the twin-twin transfusion syndrome includes more than the shunting of blood from donor to recipient. The pediatric definition of twin-twin transfusion syndrome does not apply in utero (8) and anemia in the donor and polycythemia in the recipient might well be confined to a subset of late-onset TTS in the third trimester. Imbalance in the net flow of blood across the placental vascular communications from one fetus, the donor, to the other, the recipient, could result from primary placental insufficiency on the donor side (9). Subsequent increased peripheral resistance in the donor would promote shunting of blood to the recipient; the donor therefore suffers from both hypovolemia due to blood loss and hypoxia due to placental insufficiency. The recipient fetus compensates for its expanded blood volume with polyuria (10), but, since protein and cellular components remain in the circulation of the fetus, the consequent increase in colloid pressure could draw water from the maternal compartment across the placenta. Indeed, there are larger inter-twin differences in concentrations of plasma proteins, such as albumin and total protein, in pregnancies complicated by TTS than in control twins (11). A vicious circle of hypervolemia-polyuria-hyperosmolality is established, leading to polyhydramnios and high output heart failure, with consequent spontaneous miscarriage or very premature delivery. Cardiac overload and dilatation in the recipient could promote high atriopeptin secretion by the dilated atria and increase further these vascular changes; they might then become an autonomous phenomenon that could potentially continue after interruption of the feto-fetal transfusion (12). This paper suggests that this could occur very early in the gestation, as a consequence of early cardiac overload, and trigger TTS. Sebire and colleagues (2) present their results as a potential screening test for TTS. A screening policy is usually considered to be justified when: (1) The condition can be diagnosed by definite criteria There is still little agreement on the definition of TTS. The most commonly accepted criterion to describe twin-twin transfusion syndrome presenting in the mid-trimester is the discordancy in amniotic fluid volume. This can be very inaccurate unless oliguria and oligohydramnios develop in the donor, and polyuria with subsequent severe polyhydramnios (deepest pool > 8 cm) can be demonstrated in the recipient. The discrepancy in size is frequent but not mandatory; both fetuses are of the same sex and the pregnancy might be known to be monochorionic. Although placental mapping of the anastomotic vessels does not seem clinically relevant (13), the demonstration of cardiac overload or congestive heart failure in a recipient twin should be an essential diagnostic criterion of the syndrome (14). (2) The condition screened for is prevalent and the severity of the disease leads to a poor outcome with expectant management Twin-twin transfusion syndrome complicates 15% (4-35%) of monochorionic twin pregnancies, representing, therefore, around 1 % of all twin pregnancies (15,16) and accounts for l7% of perinatal mortality (17), nearly 12% of neonatal deaths and 8.4% of infant deaths in twins. This is 3-10 times higher than that attributed to singletons (15). This increased mortality is 2-3 times higher in monochorionic twins, as compared to dichorionic twins. Neurodevelopmental abnormalities are 6-8 times more frequent in twins than in singletons. This is mainly related to prematurity and low birth weight; however, the increased risk to twins of cerebral palsy is not entirely explained by their increased risk of prematurity and low birth weight (19). Comparisons of twin pairs of like sex and unlike sex consistently show higher perinatal mortality rates in the like-sex pairs, and the increase in perinatal mortality appears to be confined to monochorionic pairs (3) Prenatal treatment reduces the perinatal mortality and morbidity associated with expectant management Expectant management in TTS presenting before 28 weeks is associated with 80-100% perinatal mortality (15,21). The dismal prognosis has generated enthusiasm for various invasive management procedures, such as serial amniodrainage (22) and laser ablation of placental anastomoses (23,24) which have resulted in significant improvement of this condition. However, both therapeutic approaches are likely to represent only attempts to treat the symptoms rather than the underlying condition. In contrast to endoscopic surgery, where survival both with time and in different centers remains stable at 55% for both fetuses and more than 70% for pregnancies with at least one survivor (24), with serial amniodrainage, there are marked differences in reported results from different studies. As reported by Saunders and colleagues (22), in studies before 1991, the survival for fetuses was 30-40%, whereas in the last 6 years, using apparently the same technique, the survival has improved up to 83% (25). Thus, three centers featuring in five different papers (21,25-28) have reported that, in 53 pregnancies treated with serial amniocentesis, and in 14 managed with watchful waiting, the perinatal survival rates were 76% and 36%, respectively suggesting that the condition was milder than shown in previous studies. Also of importance, the largest series included six dichorionic pregnancies The latest publication on the topic (29) reports an overall neonatal survival rate of 57% in 26 pregnancies treated by serial amniocentesis, which is less optimistic than the other series. (4) The antenatal intervention is associated with a reduction in the perinatal morbidity. Pinette and colleagues reported that 36% of the survivors in a population treated by serial amniocentesis had cerebral palsy (25). Bajoria and colleagues noted the presence of porencephalic cysts in 29% of the survivors (four out of 14) and cardiac dysfunction in another four survivors (30). Trespidi and colleagues reported a severe handicap rate of 15% in the surviving twins (four out of 26), when managed similarly (29). Fusi and co-workers found in four out of seven (57%) monochorionic diamniotic pregnancies that the survivors had serious problems following the demise of their co-twin (31). Liu and colleagues reported similar figures - 13 out of 25 (52%) (32). Intrauterine death of one twin and twin-twin transfusion syndrome are associated with a further 3-10 times increase in neurological morbidity (53). Visceral damage following vascular accidents, mainly cerebral and renal, are the main factors causing morbidity for the survivor of a monochorionic twin pregnancy (34, 35). In one study, neurological morbidity was detected in 26% of surviving twins and their mortality was 12%. The hypothesis that there is a passage of thromboplastin-like substances from the dead twin through placental vascular anastomoses to the surviving twin, leading to intrauterine disseminated intravascular coagulation (37,55) has never been substantiated by fetal or neonatal investigation. The death of one twin in utero is not an exclusive situation predisposing to cerebral palsy in monochorionic twins. Severe hemodynamic changes, as well as prolonged hypoxia or cerebral edema in hydropic fetuses, can lead to cerebral necrosis when both twins are delivered alive (39, 40). Although the role, and even the existence of the placental anastomoses, have been recently challenged (40), they have been found constantly in monochorionic placentas for the last 100 years (41), and, irrespective of their nature (artery-artery, vein-vein or artery-vein), and their potential role in the pathophysiology of TTS, the anastomotic branches will run on the surface of the placenta and join either superficially or in the depth of a placental unit. Any vessel joining the two fetal circulations would act as a syphoning system from the twin with the higher blood pressure towards the twin with the lower blood pressure. This morbid effect is maximal at the time of the intrauterine death of one twin, where the survivor exsanguinates into the dead twin. In an ongoing series of over 130 cases treated with laser coagulation of the placental anastuinoses, we have found an incidence of neurological impairment of less than 5% in the survivors with at least a 12-month follow-up (5) The ideal screening test is easy to perform, cheap and non-invasive and the sensitivity of the test is high Measurement of nuchal translucency thickness at 11-14 weeks is easily performed by trained sonographers and is now widely performed in screening programs for fetal aneuploidy. A nuchal translucency measured above the 95th centile is readily recognizable but, as this could be found in only 30% of severe TTS (2), this sign is unlikely to significantly enhance the prediction of TTS. Instead, the authors suggest examining the inter-twin difference in nuchal translucency measurements to determine those which are going to develop TTS. However, the cut-off in delta nuchal translucency was 0.7 mm, which is unlikely to be clinically very relevant for improving the detection rate of TTS. The first trimester of pregnancy is the period of choice for determination of the chorionicity of a twin pregnancy because the constituent layers of the inter-twin membrane can be recognized ultrasonographically (42). Fusion of the chorion and amnion, starting at 11 weeks, would leave a gestation window up to 14 weeks to detect the presence or absence of the lambda sign which is the thick attachment of a four-layer membrane to the placenta (43). Since around 15% of monochorionic twin pregnancies are likely to develop twin-twin transfusion syndrome, and these represent approximately 1/400 pregnancies, the high-risk group should be easily identified, and the complication diagnosed early, by fortnightly ultrasound examination looking for one or more ultrasound features of TTS. Early recognition of the syndrome could lead to early inter-vention, before irreversible damage is caused to one or both twins. A randomized study will soon compare serial amniodrainage and laser coagulation of the placental vessels (44). This technique should be carefully evaluated before turning towards alternative methods with uncertain rationale and few preliminary data such as septostomy (45). Eighty patients in each group will be necessary to show a significant difference in perinatal mortality and morbidity.  Fetal Medicine Unit St. George's Hospital Medical School London, UK  REFERENCES 1. The history of Isaac and his sons. Genesis, 25, 19-36 2. Sebire, N.J., D'Ereole, C., Hugues, K., Carvalho, M. and Nicolaides, K. H. (1997). Increased nuchal translucency thickness at 10-14 weeks of gestation as a predictor of severe twin-to-twin transfusion syndrome. Ultrasound Obstet. Gynecol., 10, 86-9 3. Burn, J. and Corney, G. (1984). Congenital heart defects and twinning. Acta GeneL Med. Gemello. Roma, 33, 61-9 4. Popeck, E. J., Strain, J. D., Neumann, A. and Wilson, H. (1993). In utero development of pulmonary artery calcification in monochorionic twins: a report of three cases and discussion of the possible aetiology. Pediatr. Pathol., 13, 597-611 5. Zosmer, N., Barjoria, R., Weiner, E. and Fisk, N. (1994). Clinical and echographie features of in utero cardiac dysfunction in the recipient twin in twin-to-twin transfusion syndrome. Br. Heart., 72, 74-9 6. Hecher, K., Sullivan,J. D. and Nicolaides, K. H. )1994). Temporary iatrogenic fetal tricuspid atresia in a case of twin-to-twin transfusion syndrome. Br. Heart J., 72, 457-60 7. Tanaka, M., Natori, M., Ishimoto, H., Kohno, H., Kol'ayashi, T. and Nosawa, S. 1992). Intravascular pancuronium bromide infusion for prenatal diagnosis of twin-to-twin transfusion syndrome. Fetal Diagn. Ther., 7, 3640 8. Fisk, N.M., Borrell, A., Hubinont, C., Tannirandorn, Y., Nicolini, U. and Rodeck, C. H. (1990). Fetofetal transfusion syndrome: do the neonatal criteria apply in utero? Arch. Dis. Child., 65, 657-61 9. Saunders, N.J., Snijders, R. j. M. and Nicolaides, K. H. (1991). Twin-twin transfusion syndrome during the second trimester is associated with small intertwin differences. Fetal Diagn. Ther., 6, 34-6 10. Rosen, D., Rabinowitz, R., Beyth, Y., Feijgin, M. and Nicolaides, K. H. )1990). Urine production in normal twins and in twins with acute polyhydramnios. Fetal Diagn.,5, 57-60 11. Berry, S. M., Puder, K. S., Bottoms, S. F., Uckele, J. F., Romero, R. and Cotton, D. B. (1995). Comparison of intra-uterine hematological and biochemical values between twin pairs with and without stuck twin syndrome. Am.]. Obstet. Gynecol., 172, 1403-10 12. Nageotte, M. P., Hurwitz, S. R., Vaziri, N. D. and Pandian, M. R. (1989). Atriopeptin in the twin transfusion syndrome. Obstet. Gynecol., 73, 867-70 13. Hecher, K., Ville, Y. and Nicolaides, K. H. (1995). Color Doppler ultrasonography in the identification of communicating vessels in twin-twin transfusion syndrome and acardiac twins.]. Ultrasound Med., 14, 3740 14. Hecher, K., Ville, Y., Snijders, R. and Nicolaides, K. H. (1995). Doppler studies of the fetal circulation in twin-twin transfusion syndrome. Ultrasound Obstet. Gynecol., 5, 318-24 15. Patten, R. M., Mack, L. A., Harvey, D., Cyr, D. R. and Pretorius, D. H. (1989). Disparity of amniotic fluid volume and fetal size: problem of the stuck twin - US studies. Radiology, 172, 153-7 16. Robertson, E.G. and Neer, K. J. (1983). Placental injection in twin gestation. Am.]. Obstet. Gynecol., 147, 1704 17. Steinberg, L. H., Hurley, V. A., Desmedt, F. and Beiseber, N. A. (1990). Acute polybydramnios in twin pregnancies. Aust. N.Z. J. Obstet. Gynecol., 30, 196-200 18. Powers, W. F. and Kiely, J. L. (1994). The risks confronting twins: a national perspective. Am. J. Obstet. Gynecol., 170, 19. Williams, K., Hennessy, F. and Alberman, F. 1996). Cerebral palsy: effects of twinning, birthweigbt and gestational age. Arch. Dis. Child., 75, Fl 78-82 20. Rydbstrom, H. (1994). Discordant birthweight and late fetal death in like-sexed and unlike-sexed twin pairs: a population-based study. Br.]. Obstet. Gynaecol., 101, 765-9 21. Mahony, B. S., Petty, C. N., Nyberg, D. A., Luthy, D. A. and Hirsch, J. H. (1990). The stuck twin phenomenon: ultra- sonographic findings, pregnancy outcome and management with serial amniocenteses. Am. ]. Obstet. Gynecol., 163, 1513-22 22. Saunders, N.J., Snijders, R. J. M. and Nicolaides, K. H. (1992). Therapeutic amniocentesis in twin-twin transfusion syndrome appearing in the second trimester of pregnancy. Am.]. Obstet. Gynecol., 166, 8204 23. DeLia, J. F., Kuhlmann, R. S., Harstad, T. W. and Cruishank, D. P. (1995). Fetoscopic laser ahlation of placental vessels in severe previable twin-twin transfusion syndrome. Am. J Obstet. GynecoL, 172, 1202-11 24. Ville, Y., Hyett, J., Hecher, K. and Nicolaides, K. H. (1995). Preliminary experience with endoscopic laser surgery for severe twin-twin transfusion syndrome. N. EngI.]. Med., 332, 224-7 25. Pinette, M. G., Pan, Y., Pinette, S. G. and Stubblefield, P. G. (1993). Treatment of twin-twin transfusion syndrome. Obstet. Gynecol., 163, 1513-22 26. Urig, M. A., Clewel., W. H. and Elliott, J. P. (1990). Twin-twin transfusion syndrome. Am. ]. Obstet. Gynecol., 163, 1522-6 27. ElluIrt, j. P., Urig, M. A. and Clewell, W. H. (1991). Aggressive therapeutic amnulcentesis for treatment of twin-twin transfusion syndrome. Obstet. Gynecol., 77, 537~0 28. Reisner, D. P., Mahony, B. S. and Petty, C. N. (1993). Stuck twin syndrome: outcome in thirty-seven consecutive cases. Am.]. Obstet. Gynecol., 169, 991-5 29. Trespidi, L., Bosehetto, C., Caravelli, F., Villa, L., Klustermann, A. and Nicolini, U. (1997). Serial amniocentesis in the management of twin-twin transfusion syndrome: when is it valuable? Fetal Diagn. Ther., 12, 15-20 30. Baulna, R., Wigglesworth, J. and Fisk, N. (1995). Angioarcbitecture of monoeborionic placentas in relation to the twin-twin transfusion syndrome. Am. ]. Obstet. Gynecol. 172,856-63 31. Fusi, I.. and Gordon, H. (1 990). Twin pregnancy complicated by intrauterine death. Problem and outcome with Conservative management. Br.]. Obstet. Gynaecol., 97,511-16 32. Liu, S., Bernirschke, K., Scioscia, A. L. and Mannino, F. L. (1992). Intrauterine death in multiple gestation. Acta Genet. Med. GemelloL, 41, 5-26 33. Petterson, B., Nelson, K. B., Watson, L. and Stanley, F. (1993). Twins, triplets, and cerebral palsy in births in Western Australia in the 1 980s. Br. Med.]., 307, 1239~3 34. Isimatu, J., Hon, D., Miyajima, S., Hamada, T., Yakusbiji, M. and Nishimi, T. (1994). Twin pregnancies complicated by the death Of one fetus in the second or third trimester.]. Matern. fetal Invest., 4, 141 35. Prompeler, H.]., Madjar, H., Kiosa, W., Dubois, A., Zabradnik, H. P., Schillinger, H. and Breckwoldt, M. (1994). Twin pregnancies with single fetal death. Acta Obstet. Gynecol Scand., 73, 205-8 36. Okamura, K., Murorsuki, J., Tanigawara, S., Uehara, S, and Yajima, A. (1994). Funipuncture for evaluation of hematologic and coagulation indices in the surviving twin following co-twin's deatb. Obstet. Gynecol., 83, 975-8 37. Benirschke, K. (1961). Twin placenta in perinatal mortality. N.Y. State]. Med., 61, 1499-508 38. Moore, M., McAdams, A. and Sutherland, J. (1969). Intrauterine disseminated intravascular coagulation: a syndrome of multiple pregnancy with a dead fetus.]. Pediatr., 74, 523-8 39. Bejar, R., Vigliocco, G. and Gramajo, H. (1990). Antenatal origin of neurological damage in newborn infants in Multiple gestations. Am.]. Obstet. Gynecol., 162, 1230-6 40. Larroche, J. C., Droolle, P., Delezoide, A. L., Narcy, F. and Nessman, C. (1990). Brain damage in monozygous twins. BirIL Neonate, 57, 261-78 41. Schatz, F. (1882). Fine besondere art von elseitiger polybydramnie mit anderseirger oligohydramnie bei emugen zwillingen. Arch. Gynaekol., 19, 329-69 42. Monteagudo, A., Timor-Tritsch, I. and Sharnia, 5. (1994). Early and simple determination of chorinnic and ainnioric type in multifetal gestations in the first 14 weeks by high frequency transvaginal ultrasonography. Am. j. Obstet. Gyneci)L, 170, 824-9 43. Bessis, R. and Papiernik, F. (1981). Echographic imagery of amniotic membranes in twin pregnancies. Twin Research 3: Twin Biology and Multiple Pregnancy, pp.183-87. (Alan Liss Inc.) 44. Eurofetus. European Working Group on Endoscopic Access to the Fetoplacental Unit. Biomed 2 Programme ot the European Commission 45. Saade, G. R., Olson, G., Belfori, M. A. and Moise, K. (1995). Amniotomy: a new approach to the stuck twin syndrome. Am.]. Obstet. Gynecol., 172, 429