Outcome of twin pregnancies complicated by single intrauterine death in relation to vascular anatomy of the monochorionic placenta

Rekha Bajoria1, Ling Y. Wee, Shaheen Anwar and Stuart Ward

University of Manchester, St Mary's Hospital, Department of Obstetrics & Gynaecology, Whitworth Park, Manchester M13 OJH, UK


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The objective of this study was to determine the relationship between the type of placentation, vascular anatomy of the monochorionic (MC) placenta and the perinatal outcome of the surviving twin following a single intrauterine fetal death (IUFD). In this retrospective study, 92 twin pregnancies complicated by a single intrauterine death were identified from three tertiary referral centres [50 MC and 42 dichorionic (DC)]. Antenatal and neonatal data as well as information on the chorionicity, vascular anastomoses, and autopsy findings were also obtained. The percentage risk of IUFD (26 versus 2.4; P < 0.001), anaemia (51.4 versus 0; P < 0.001) and intracranial lesions at birth (46 versus 0; P < 0.001) was greater in MC than in DC twins. In MC twins without twin–twin transfusion syndrome (TTTS), perinatal mortality was higher in the group with superficial arterioarterial (AA)/venovenous (VV) channels than those with only multiple bidirectional arteriovenous (AV) anastomoses (12/15 versus 0/8; P < 0.001). However, in the TTTS pregnancies (n = 26), perinatal outcome of the surviving twin was dependent on whether the recipient (n = 16) or the donor twin (n = 10) died first. Incidence of IUFD (9/16 versus 0/10; P < 0.001), severe anaemia (7/7 versus 1/10; P < 0.001) and intracranial lesions at birth (6/7 versus 2/10; P < 0.001) was common in pregnancies where the recipient twin died first. In the TTTS group, unidirectional AV anastomotic channels were found in all but two placentae. In conclusion, this study suggests that the outcome of twin pregnancies complicated by IUFD is dependent on the type of vascular anastomoses and TTTS.

Key words: co-twin demise/monochorionic multiple pregnancy/placental vascular anastomoses/polyhydramnios


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Antepartum death of a single fetus complicates 2.5–5% of twin pregnancies and may be associated with a significant morbidity and mortality in the surviving co-twin (Gaucherand et al., 1994Go; Murphy, 1995Go; Bajoria and Kingdom, 1997Go). In monochorionic (MC) pregnancies the perinatal loss rates and risk of death or neurological handicap in the surviving co-twin is 3–4-fold greater than for dichorionic (DC) pregnancies (Enbom, 1985Go; Benirschke, 1993Go). However, regardless of the chorionic status, none of these complications occurs in pregnancy with the vanishing twin (Landy and Keith, 1998Go).

The precise cause of poor perinatal outcome of the surviving co-twin in MC pregnancies is unknown. Acute haemodynamic imbalance resulting from anastomotic flow is considered to be the likely mechanism (Fusi and Gordon, 1990Go; Fusi et al., 1991Go). Although the chorionic plate vascular anatomy of MC placenta was first described more than a century ago (Schatz, 1890Go), no studies to date have made an attempt to relate the vascular anatomy to the perinatal mortality and morbidity of the surviving co-twin. Many clinical studies have considered the presence of superficial vascular anastomoses to be a risk factor for the surviving twin (Bejar et al., 1990Go; Fusi et al., 1991Go) but have failed to provide comparative data on anastomoses and favourable outcome. Anecdotal data suggest that the risk of occurrence of co-twin sequelae is 3-fold greater in MC pregnancies complicated by twin–twin transfusion syndrome (TTTS) (Arts et al., 1996Go). Recently studies have shown that TTTS is caused by the presence of a unidirectional deep arteriovenous (AV) shunt with paucity of superficial anastomoses (Bajoria et al., 1995Go; Machin et al., 1996Go; Bajoria, 1998bGo). Furthermore, it has been suggested that it is the superficial anastomoses which are responsible for acute transfusional complications following intrauterine fetal death (IUFD) of one of the twins (Fusi et al., 1991Go). Taken together, this evidence raises the possibility that the risk to the surviving MC co-twin may depend upon the type and the size of the vascular shunts, rather than the presence or absence of superficial anastomoses.

The aim of this study was to establish if there was an association between placental vascular anatomy of twin pregnancies complicated by a single intrauterine death and the perinatal outcome of the surviving co-twin. In addition, in TTTS pregnancies an attempt was also made to determine the neonatal outcome of the surviving twin in relation to whether antepartum death of the donor or the recipient twin occurred first. This information is important as it may influence the clinical management of this unusual and difficult problem, especially since delineation of superficial vascular anastomoses by power Doppler has now become a realistic possibility (Fortunato, 1996Go; Haberman et al., 1997Go).


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
This retrospective study was conducted at Queen Charlotte's Hospital, London, UK (1980–1995), Hammersmith Hospital, London, UK (1980–1997) and St Mary's Hospital, Manchester, UK (1991–1998). Twin pregnancies complicated by intrauterine death of one twin were identified using a triple system of data collection. Mothers who underwent abortion, had acardiac twins or who delivered before 20 weeks of gestation were excluded from the study. Twin gestations complicated by antepartum death were identified from the delivery, neonatal and perinatal pathology log books. The mother's antenatal record, neonatal chart, and neonatal discharge summaries were reviewed. Autopsy and histology reports were also obtained. Information relating to a range of antenatal, intrapartum and neonatal complications was also retrieved. The approximate time interval between the death of the first twin and the delivery of the surviving co-twin was also calculated from information recorded in the case notes.

Information on associated maternal complications such as pre-eclampsia, diabetes and fetal wellbeing were collected. Ultrasound scan data particularly in relation to chorionicity, fetal growth, liquor volume and diagnosis of TTTS were obtained. The diagnosis of TTTS was made subject to the following criteria: (i) monochorionic placentation; (ii) discordance in estimated birthweight of >15%; and (iii) oligohydramnios in the smaller and polyhydramnios in the larger twin sac. Management details of TTTS pregnancies were also obtained. Women with TTTS were managed by serial amnioreduction. One patient who underwent laser ablation of the anastomosing placental surface vessels was excluded from the study.

Following IUFD, the decision when to deliver the surviving twin was at the discretion of the consultant in charge. In mothers treated conservatively, fetal growth and wellbeing were monitored by ultrasound scans, biophysical profiles and Doppler studies. Elective delivery was only undertaken if there was a risk of fetal compromise, threatened preterm labour or chorio-amnionitis.

In the surviving twins, neonatal information at birth on haemoglobin, blood transfusion, and cranial ultrasound scan findings were obtained. The long-term follow-up data in relation to neurological handicap were obtained from the neonatal notes. In this study, only information on major neurological handicaps such as cerebral palsy, blindness and paralysis was obtained.

Autopsy data on twins who died in utero or in the early neonatal period were also collected in relation to intrauterine growth restriction, hydrops, pallor, congestion and cardiomegaly. Histological information on the placenta such as chorionicity, marginal (within 1 cm of placental margin) or velamentous cord insertion, and the type of vascular anastomotic channels was obtained.

Statistical analysis
All results were expressed as median and range. Two groups were compared by the Mann–Whitney test for continuous variables and by Fisher's exact test for blocked variables.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
During the study period, 101 twin pregnancies complicated by a single IUFD were identified. Four monoamniotic and five diamniotic (DA) pregnancies were excluded because information on the type of anastomoses and/or neonatal data were not available. Perinatal outcome data on 92 twin pregnancies in relation to chorionicity, TTTS and type of vascular anastomoses are reported.

Comparison between MC and DC twin pregnancies (Table IGo)
The median gestational age of first intrauterine death in MC pregnancies (n = 50) was comparable to that of DC pregnancies. However, the median gestational age at delivery of the second twin in MC was lower than in DC twins (P < 0.002). In MC twins, incidence of IUFD of the co-twin (13/50 versus 1/42; P < 0.001), and total perinatal mortality rate (29/50 versus 9/42; P < 0.01) were higher than DC twins. In the MC group, anaemia was found in 19/37 of the surviving co-twins while polycythaemia with a haematocrit of 95 was present in only one twin. In contrast, none of the surviving co-twins in the DC group required blood transfusion. Prior to intrauterine demise of the first twin, discordance in estimated birthweight was more common in MC than DC twins (39/50 versus 13/42; P < 0.01).


View this table:
[in this window]
[in a new window]
 
Table I. Outcome of the surviving co-twin in relation to chorionicity
 
Comparison in MCDA twins with or without TTTS (Tables II and IIIGoGo)
When the outcomes of pregnancies in MC twins were analysed in relation to TTTS, the median (range) gestational age of death of one twin was greater in non-TTTS than TTTS pregnancies [28.5 (21.5–38) versus 25.2 (22.3–34) weeks; P < 0.0001]. Similar differences were also found in gestational age at delivery [30 (21.6–38) versus 27.1 (22.3–35) weeks; P < 0.009]. Incidence of intrauterine death of the second twin (4/24 versus 9/26), and total perinatal mortality (12/24 versus 17/26) were similar in the two groups. Similarly, the incidence of anaemia (11/20 versus 8/17), significant abnormality on the intracranial scan (15/20 versus 14/17) were not significantly different between the two groups. Histology confirmed MCDA placentation. In the non-TTTS group (Table IIGo), 15 placentae had superficial anastomoses of arterioarterial (AA)/venovenous (VV) type, whereas nine had only multiple deep arteriovenous (AV)/venoarterial (VA) anastomoses. In contrast, 24 placentae with TTTS had only deep AV anastomoses, whereas additional superficial anastomoses were found in only two placentae. The frequency of abnormal cord insertion between the two groups was comparable. Individual data of both the groups are given in Tables II and IIIGoGo.


View this table:
[in this window]
[in a new window]
 
Table II. Perinatal outcome of co-twins after a single fetal death in 24 monochorionic twin pregnancies without twin–twin transfusion syndrome
 

View this table:
[in this window]
[in a new window]
 
Table III. Perinatal outcome of co-twins after a single fetal death in 26 monochorionic twin pregnancies without twin–twin transfusion syndrome
 
In the non-TTTS pregnancies, incidence of intrauterine death (4/15 versus 0/9; not significant), and among live births, incidence of anaemia (10/11 versus 1/9; P < 0.05), intracranial lesions (9/11 versus 2/9; P < 0.01) and neurological handicap in the surviving co-twin (i.e. excluding IUFD and neonatal death 2/3 versus 0/9; P < 0.05) were more common in the presence of superficial placental anastomoses compared with multiple AV/VA anastomoses.

TTTS pregnancies in relation to demise of the recipient or donor twin (Table IIIGo)
In 26 cases of pregnancies complicated by chronic mid-trimester TTTS, data sets were divided into two subsets according to whether the recipient or the donor twin died first. In 16 women, the recipient twin died first while in 10 cases, donor twin death occurred first.

The median (range) gestational age of intrauterine death of the first twin was comparable between the two groups [25 (22.3–34) versus 28.4 (23–30) weeks]. The median gestational age at delivery in the group where the donor twin died first was greater than those pregnancies where the recipient twin died first [29.3 (26.2–34.2) versus 25 (22.3–34.1); P < 0.004]. Similarly, intrauterine death (9/16 versus 0/10; P < 0.001), and total perinatal mortality rate (15/16 versus 2/10; P < 0.001) were higher with first death of the recipient than the donor twin. Frequency of anaemia requiring blood transfusion was also higher in the group where the recipient twin died first (excluding IUFD) than when death of the donor twin occurred first (7/7 versus 1/10; P < 0.001). Examination of the placenta confirmed monochorionicity. The frequency of abnormal cord insertion was comparable between two groups. The chorionic plate vascular anatomy between the groups was comparable in terms of presence of AV anastomoses with or without superficial AA/VV shunts. Intrauterine demise of the donor was associated with spontaneous resolution of polyhydramnios and hydrops in 6/10 recipient twins. The perinatal outcome in relation to vascular anatomy is summarized in Figure 1.Go



View larger version (24K):
[in this window]
[in a new window]
 
Figure 1. Illustration of the perinatal outcome of twin pregnancies complicated by a single intrauterine death in relation to vascular anatomy of the placenta.

 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
This study shows that the perinatal outcome of a surviving twin depends upon the type of placentation and confirms the notion that the risk of neurological handicap/death in the co-twin is greater in MC than DC pregnancies (Murphy, 1995Go). Although in the majority of cases the cause of the first intrauterine death remained indeterminate, discordant fetal growth, polyhdramnios–oligohydramnios, and abnormal cord insertion were more common in MC than DC twins.

Traditionally, it has been believed that the demise/neurological lesion of the co-twin is due to the passage of thrombotic or necrotic materials from the dead to the healthy twin along placental vascular shunts (Benirschke, 1993Go). The validity of this idea has been questioned recently because of the normal coagulation status and anaemia in the survivors (Okamura et al., 1994Go). The data reported in this study provide a morphological basis to the haemodynamic imbalance theory which states that placental anastomoses allow transfer of blood from the surviving twin to the dead co-twin, giving rise to periods of hypoperfusion resulting in neurological changes (Bajoria and Kingdom, 1997Go). Therefore, transfusion of biologically active compounds from the dead to the surviving twin is unlikely.

We found a normal outcome in the surviving recipient twin following the sudden demise of the donor twin. All except one recipient twin had a normal haemoglobin at birth with no adverse neurological outcome. Furthermore, resolution of polyhydramnios and hydrops in the majority of the recipient twins suggests that neither the transfusion of blood nor the passage of acute thrombotic material between the fetuses took place. This favourable outcome can be explained on the basis of the type of the anastomotic channels present in the feto-placental unit. All except one placenta had AV anastomoses connecting the arterial circulation of the donor to the venous end of the recipient twin. It has previously been shown, using controlled perfusion experiments, that AV shunts allow unidirectional flow, i.e. from the artery to vein (Bajoria et al., 1995Go; Bajoria, 1998aGo,bGo). It is therefore conceivable that a favourable outcome for the recipient twin might be attributed to no flow between the surviving and dead twins simply because of the negative pressure gradient along the AV anastomoses (dead donor with zero and alive recipient twin with higher systemic pressure).

In keeping with this proposition, data from this study also show that in placentae with unidirectional AV anastomoses, death of the recipient twin was invariably associated with intrauterine or neonatal death of the donor twin. As all the seven donor twins born alive were severely anaemic and required blood transfusion, transfer of blood along the unidirectional anastomoses seems to be the most likely explanation for such a dismal outcome in this group. The direction of the blood flow through AV anastomoses argues against the possibility of transfer of thrombotic material from the dead to the live donor twin. Furthermore, formation of thrombotic material is unlikely because of the temporal proximity of the sequelae to the initial recipient twin death (~48 h). Consistent with this proposition, autopsy finding on twin pairs dying in utero revealed pallor in the donor and polycythaemia in the recipient twin.

Hence, from the clinical perspective, these data suggest that in the event of intrauterine demise of the recipient twin, the donor can only be rescued by prompt delivery after 28 weeks gestation. However, in pregnancies where both twins are still viable, an alternative option such as the occlusion of the umbilical cord of the recipient twin is a distinct possibility (Bebbington et al., 1995Go; Deprest et al., 1996Go).

We also observed a higher incidence of co-twin sequelae in terms of intrauterine death, neonatal anaemia, abnormal intracranial ultrasound scan, and neurological handicap in the presence of superficial AA/VV anastomoses. In our cohort only one of the placentae had VV anastomoses in isolation. The severity of anaemia and the autopsy findings from twin pairs who died in utero further lend credence to the haemodynamic theory. It is conceivable that in the presence of superficial AA anastomoses, a massive transfer of blood can occur, against the pressure gradient from the live to the dead twin. This may cause brain damage or fetal demise of the surviving twin simply because of severe haemodynamic imbalance. However, transfusion of thrombotic material from the dead to the live fetus is virtually impossible because the pressure gradient is such that flow from the dead to the live fetus cannot take place.

A favourable neonatal outcome was seen in MC pregnancies with multiple bidirectional AV anastomoses without superficial anastomoses. None of the twins in this category had significant anaemia at birth and had normal neurodevelopment. On the other hand, one baby who was born alive had severe polycythaemia (haematocrit of 94) and required an exchange transfusion. The optimal outcome of MC twins can only be explained on the basis of attainment of a dynamic steady state along the AV and VA anastomotic channels with oppositely directed anastomotic blood flow. We speculate that in this group, following the demise of one twin, a massive blood transfusion must have occurred from the survivor's arterial to the dead twin's venous circulation (AV). In the dead twin, this may lead to a rise in the systemic filling pressure which then in turn can initiate the flow along the VA anastomoses with the establishment of an intertwin circulation. In this cohort, it is theoretically possible that thrombotic material generated in the dead twin may reach the circulation of the viable fetus. However, the normal outcome in eight cases with no evidence of neurological handicap argues against the thromboembolic episode as the cause for co-twin sequelae.

The retrospective nature of this study means that information relating to the vascular anastomoses may be incomplete. We accept that delineation of vascular shunts by conventional techniques may not be as accurate and precise as reported previously where controlled perfusion experiments were undertaken (Bajoria, 1998bGo). Nevertheless, identification of superficial channels by injection technique is quite simple and does not require a high degree of skill. Furthermore, in this study we evaluated, in the first instance, the data according to the presence or absence of superficial anastomoses. Only in the group without superficial anastomoses were direction and numbers of AV anastomoses taken into account. The data relating to TTTS placentae generated by a conventional injection method in this study agrees, at least in principle, with the previously published data where physiological techniques were used (Bajoria et al., 1995Go; Bajoria, 1998aGo,bGo). Furthermore, the primary objectives of this study were to evaluate the relationship between the co-twin sequelae and the type of shunt. Further prospective study is necessary to confirm these findings.

In conclusion, our data suggest that the co-twin sequelae are dependent upon the type of chorionic plate vascular anatomy. Development of newer techniques to map the vascular anatomy of MC placentae accurately in vivo are therefore warranted as the availability of such information during the antenatal period is likely to influence the clinical management and prevent neurological handicap in at least some of the surviving co-twins.


    Acknowledgments
 
We are grateful to the staff of Perinatal Pathology, Centre of Fetal Care, Labour Ward and Neonatal Unit of Queen Charlotte's and Hammersmith Hospital for their assistance in the collection of the clinical data.


    Notes
 
1 To whom correspondence ahould be addressed Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Arts, H., van Eyck, J. and Arabin, B. (1996) Fetal death of one twin in a monochorionic pregnancy with twin–twin transfusion syndrome. J. Reprod. Med., 41, 775–778.[ISI][Medline]

Bajoria, R. (1998a) Abundant vascular anastomoses in monoamniotic versus diamniotic monochorionic placentas. Am. J. Obstet. Gynecol., 179, 788–793.[ISI][Medline]

Bajoria, R. (1998b) Vascular anatomy of monochorionic placenta in relation to discordant growth and amniotic fluid volume. Hum. Reprod., 13, 2933–2940.[Abstract/Free Full Text]

Bajoria, R. and Kingdom, J. (1997) A case for routine determination of chorionicity and zygosity in multiple pregnancies. Prenat. Diagn., 17, 1207–1225.[ISI][Medline]

Bajoria, R., Wigglesworth, J. and Fisk, N.M. (1995) Angioarchitecture of monochorionic placentas in relation to the twin–twin transfusion syndrome. Am. J. Obstet. Gynecol., 172, 856–863.[ISI][Medline]

Bebbington, M.W., Wilson, R.D. Machan, L. and Wittmann, B.K. (1995) Selective feticide in twin transfusion syndrome using ultrasound-guided insertion of thrombogenic coils. Fetal Diagn. Ther., 10, 32–36.[ISI][Medline]

Bejar, R., Vigliocco, G., Gramajo, H. et al. (1990) Antenatal origin of neurologic damage in newborn infants. II. Multiple gestations. Am. J. Obstet. Gynecol., 162, 1230–236.[ISI][Medline]

Benirschke, K. (1993) Intrauterine death of a twin: mechanisms, implications for surviving twin, and placental pathology. Semin. Diagn. Pathol., 10, 222–231.[ISI][Medline]

Deprest, J.A., Evrard, V.A., Van Schoubroeck, D. and Vandenberghe, K. (1996) Endoscopic cord ligation in selective feticide. [Letter.] Lancet, 348, 890–891.

Enbom, J.A. (1985) Twin pregnancy with intrauterine death of one twin. Am. J. Obstet. Gynecol., 152, 424–429.[ISI][Medline]

Fortunato, S.J. (1996) The use of power Doppler and color power angiography in fetal imaging. Am. J. Obstet. Gynecol., 174, 1828–1833.[ISI][Medline]

Fusi, L. and Gordon, H. (1990) Twin pregnancy complicated by single intrauterine death. Problems and outcome with conservative management. Br. J. Obstet. Gynaecol., 97, 511–516.[ISI][Medline]

Fusi, L., McParland, P., Fisk, N. et al. (1991) Acute twin–twin transfusion: a possible mechanism for brain-damaged survivors after intrauterine death of a monochorionic twin. Obstet. Gynecol., 78, 517–520.[Abstract]

Gaucherand, P., Rudigoz, R.C. and Piacenza, J.M. (1994) Monofetal death in multiple pregnancies: risks for the co-twin, risk factors and obstetrical management. Eur. J. Obstet. Gynecol. Reprod. Biol., 55, 111–115.[ISI][Medline]

Haberman, S., Haratz-ubinstein, N., Baxi, L. and Helle, D. (1997) Power Doppler sonography in monochoionic twins: a preliminary study. J. Matern. Fetal Invest., 7, 84–88.[ISI]

Landy, H.J. and Keith, L.G. (1998) The vanishing twin: a review. Hum. Reprod. Update, 4, 000–000.

Machin, G., Still, K. and Lalani, T. (1996 Correlations of placental vascular anatomy and clinical outcomes in 69 monochorionic twin pregnancies. Am. J. Med. Genet., 61, 229–236.

Murphy, K.W. (1995) Intrauterine death in a twin: implications for the survivor. In Ward, R.H. and Whittle, M. (eds), Multiple Pregnancy. RCOG Press, London, pp. 218–31.

Okamura, K., Murotsuki, J., Tanigawara, S. et al. (1994) Funipuncture for evaluation of hematologic and coagulation indices in the surviving twin following co-twin's death. Obstet. Gynecol., 83, 975–978.[Abstract]

Schatz, F. (1890) Die Gefässverbindungen der Placentarkreisläufe eineiiger Zwillinge, ihre Entwicklung und ihre Folgen. Arch. Gynäkol., 27, 1–72.

Submitted on February 2, 1999; accepted on April 29, 1999.