Incidence, Mechanisms, and Patterns of Fetal Cerebral Lesions in Twin-to-Twin Transfusion Syndrome

By Quarello, Edwin Molho, Marc; Ville, Yves

Abstract Objective. To determine the incidence of fetal cerebral lesions and their characteristics in twin-to-twin transfusion syndrome (TTTS).

Design and setting. This was a retrospective analysis at a single center for the period 1999 to 2004 in which 299 cases of severe TTTS at 15-28 weeks of gestation were reviewed.

Methods. Only cerebral injuries diagnosed during pregnancy or ischemic lesions diagnosed within the first week of life were considered in order to exclude those related to prematurity. We only included cases resulting in at least one survivor at one week after delivery, as well as fetuses that were terminated because of severe cerebral abnormalities. We excluded all fetuses delivered at

Results. Two hundred and ninety-nine pregnancies were evaluated. Three hundred and fifteen fetuses were reviewed. Cerebral abnormalities developed antenatally in 26/315 fetuses (8.25%). All lesions but one were diagnosed prenatally. Prenatal diagnosis of these lesions was achieved primarily by ultrasound (US) and magnetic resonance imaging (MRI), in 20/25 (80%) and in 5/25 (20%) fetuses, respectively. Cerebral abnormalities developed following primary laser coagulation in 12/222 (5.40%), following serial amnioreduction in 9/66 (13.63%), and following expectant management in 3/14 (21.4%) fetuses. Abnormalities developed after single intrauterine fetal death (IUFD) in 14 cases.

Conclusions. Cerebral morbidity in TTTS mainly occurs following vascular disruptive lesions. Both donors and recipients are at risk of developing either ischemic or hemorrhagic lesions. The risk of developing cerebral lesions in single survivors is significantly lower following laser treatment. Combined use of a targeted US and fetal MRI could detect most cerebral abnormalities antenatally. Timing of the triggering event is critical for planning serial US and MRI follow-up examinations.

Keywords: Twin-to-twin transfusion syndrome, fetal cerebral lesions, intrauterine death, hemorrhage, ischemia, ultrasound, MRI, amnioreduction, laser, ietoscopy

Introduction

The risk of developing white matter necrosis in complicated monochorionic pregnancies is 10-fold that in dichorionic twins (33% vs. 3.3%) [I]. The increased rate of prematurity and very low birdi weight in monochorionic twins cannot explain it all. Monochorionic placentation is characterized by the presence of vascular anastomoses that connect the twins’ umbilical circulations indirecdy dirough various patterns of artery-to-vein, artery-to-artery, and vein-to-vein connections on the chorionic plate. The number and distribution of chorionic plate anastomoses is likely to play a critical role in the development of twin-to-twin transfusion syndrome (TTTS) [2] in up to 15% of monochorionic pregnancy TTTS [3]. When the condition is untreated, up to 50% of single survivors are exposed to develop cerebral lesions [4] . Perinatal mortality and morbidity can originate from hemodynamic instability triggering some degree of exsanguination of one fetus into its co-twin, which culminates when one twin dies in utero.

Laser coagulation of inter-twin anastomoses is an effective surgical treatment for TTTS. The deadi of one twin occurs following serial amnioreduction in around 26% of cases and following laser coagulation in around 40% of cases [5] . Neurological morbidity in single survivors is not consistendy reported in the literature and ranges from 30% [6] to 35% [5] and from 7% [5] to 16% [7] following amnioreduction and laser coagulation, respectively.

The precise characterization of fetal cerebral lesions could improve the understanding of their padiophysiology as well as the accuracy of prenatal diagnosis. We therefore reviewed prenatal imaging of all cases of TTTS with congenital cerebral abnormalities in relation to clinical course and outcome.

Materials and methods

We reviewed all fetal cerebral lesions diagnosed in utero in TTTS cases in our unit over a five-year period Qanuary 1999 to December 2004). We only included cases resulting in at least one survivor at one week after delivery, as well as fetuses that were terminated because of severe cerebral abnormalities. We excluded all fetuses delivered at

Lesions were systematically assessed by ultrasound (US) and by magnetic resonance imaging (MRI) botii in utero and in the neonatal period. Fetal MRI examination immethately followed targeted US examination at around 32 weeks, and ultrasound findings were known by the radiologist reading the MRI images. MRI was performed earlier whenever US findings were suggestive of cerebral lesions. In the neonatal period, all scans were performed within 3 days following delivery as well as at between 4 and 7 days of life. High frequency probes were used for transabdominal (4-8 MHz), transvaginal (5-9 MHz), and transfontanellar (5-9 MHz) ultrasound examinations (GE Voluson 730 Medical Systems, Ultrasound & Primary Care Diagnostic, Gif sur Yvette, France). A 1-Tesla MRI (MRI Siemens Syngo, Erlangen, Germany) was used.

All cases were recorded and accounted for prevalence calculations. Groups were compared for cerebral abnormalities persisting over the first week of life using Fisher’s exact test. Means were compared using the Wilcoxon test. Statistical analysis was performed using Statview 4.55 software (SAS Institute Cary, NC, USA).

Screening for fetal anemia was based on sonographic features including ascites, frontal edema, hydrops, hyperechogenic bowel, and tricuspid regurgitation together with measurement of peak systolic velocity in the middle cerebral artery (MCA-PSV) [8,9] . Cerebral lesions in neonates were categorized as ischemic or hemorrhagic lesions using the classifications by Volpe and de Vries, respectively [10,11].

Results

During this five-year period, 298 pregnancies presenting with TTTS at 15-27 weeks (median 22 weeks) of gestation were managed primarily by laser coagulation (199), serial amnioreduction (79), selective feticide using cord coagulation (13), or expectant management (7). After exclusion of single and double fetal deadi and late miscarriage, 315 fetuses were reviewed, corresponding to 222, 66, 1 3, and 14 fetuses, respectively according to primary management. Two fetuses were excluded from analysis: case numbers 8 and 24 presented cerebral lesions before laser treatment.

Cerebral abnormalities were diagnosed in 26beta15 cases (8.25%). All lesions but one were diagnosed prenatally. The prevalence of fetal brain abnormalities according to primary management were of 5.40% (12/222), 13.63% (9/66), 0% (0/13), and 21.4% (3/14), respectively. MRI was not performed in 4/26 cases in which overt abnormal US features were diagnosed before 26 weeks (n = 3; numbers 10, 11, and 14) nor when delivery occurred unexpectedly at 27 weeks (n = 1; number 4) following normal US examination at 25 weeks. Antenatally, lesions were diagnosed primarily by targeted ultrasound in 20/25 fetuses (80%) and by MRI in 5/25 fetuses (20%) (Table I). MRI confirmed primary ultrasound features in 12 cases (11 abnormal and one normal). MRI features were discordant from primary abnormal US in 10 cases (five abnormal and five normal). Persistent cerebral lesions (=7 days of life) were diagnosed in 20 fetuses. Postnatal examinations were available in 16/26 fetuses (61%) either by transfontanellar US examinations in the survivors (9/16) or on postmortem examination (7/16). Postnatal examination could not be performed in cases with cerebral tissue maceration when selective feticide was performed for severe cerebral anomalies (w = 4; numbers 7, 15, 19, and 24).

The underlying persistent lesions were classified using either antenatal or postnatal imaging or by postmortem examination as being ischemic (15/20 (75%)), hemorrhagic (3/20 (15%)), or ischemichemorrhagic (2/20 (10%)). All types of lesions were evenly distributed among recipients (7) and donors (13).

Posmatal examination was at odds with prenatal ultrasound in 8/ 16 cases (50%; numbers 5, 8, 13, 18×2, 20, 22, and 23) and MRI findings in 2/16 cases (12.5%; numbers 12 and 22) (Table I).

In four pregnancies (five fetuses) ultrasound cerebral abnormalities were consistent with grade II intraventricular hemorrhage and mild ventriculomegaly (numbers 5, 8, 18 x 2, and 20) at 1 8 to 31 weeks, normalized on subsequent follow-up scans, and could not be seen on MRI 1-14 weeks later at 32 weeks of gestation nor at birth. In one fetus, pericerebral hemorrhage (number 12) diagnosed by MRI at 30 weeks was absent at birth. These five fetuses all underwent primary laser treatment.

Table I. Outcome of fetuses and neonates with cerebral lesions related to twin-to-twin transfusion syndrome.

Sixteen of the 20 persistent cerebral lesions followed serial amnioreduction (9/66) and laser coagulation (7/222). They were evenly distributed between donors (10) and recipients (6) (p = 0.003; Tables II and III). Persistent cerebral lesions were associated with the intrauterine fetal death (IUFD) of one twin in 11/21 fetuses (52.3%; five recipients and six donors) following expectant management (1/14 (7.1%)), serial amnioreduction (7/20 (35%)), and laser coagulation (3/55 (5.45%)) (? = 0.0027, Tables II and III). Cerebral injuries developed in 10 fetuses with a live co- twin. These cases included one affected twin pair (number 22) without treatment, and followed serial amnioreduction in 2/456 (4.34%), laser coagulation in 4/167 (2.39%) (p = 0.61, Tables II and III), a combined treatment in one case, and developed before a laser coagulation in one case. Bodi donors and recipients were at risk of developing either ischemic or hemorrhagic lesions (Table TV). Donors were significandy more involved following laser coagulation (p = 0.029, Table IV), especially when there were two surviving fetuses (p = 0.024, Table IV). Ten intrauterine transfusions were given to seven anemic fetuses. Transfusions were given following the death of one twin in five cases and in cases of feto-fetal hemorrhage in two cases. The mean interval between the death of one twin and the diagnosis of cerebral damage in the survivor was 5 + 3.6 weeks. This was 6.3 +- 4.2 weeks and 3.6 +- 2.5 weeks following laser and amnioreduction, respectively (p = 0.5).

Discussion

The interdependency of two fetal circulations is unique to monochorionicity. This is also the main anatomical and functional support for the development of vascular disruptive cerebral lesions. The risk culminates in single survivors but also when both twins are alive and subjected to hemodynamic instability, particularly in TTTS. Two theories have been proposed to explain the padiogenesis of these lesions over the period from the early sixties dirough to the nineties, by Benirschke [12], Bendon and Siddiqi [13], and Larroche et al. [14].

In 1961 Benirschke proposed that padiological findings in single survivors could be compatible with an embolization phenomenon, which could explain several visceral infarcts and lesions of necrosis. However, this remained only speculative and cannot account for the fact that similar brain lesions can also occur when bodi twins are born alive [I]. It is also unlikely that necrotic emboli can disseminate to the survivor against a positive pressure grathent. Bendon and Siddiqi [13] and Larroche et al. [14] highlighted the role of inter-twin hemodynamic imbalance as being the main phenomenon involved in the development of brain lesions in monochorionic twins. Aldiough a chronic and overall unidirectional transfusion from one twin, the donor, into its sibling, the recipient, merely causes significant discordance in hemoglobin levels in utero [15], hemodynamic imbalance can trigger massive feto- fetal exsanguination [4,16].

Table II. Comparison of groups with fetal cerebral lesions diagnosed within the first weeks of life according to their primary treatment and to the presence or absence of the death of one twin.

Table III. Distribution of persistent cerebral lesions according to the primary treatment and their primary characterization. Two fetuses (numbers 8 and 24) were excluded owing to the development of cerebral lesions before the primary treatment.

Table IV. Comparison of groups with fetal cerebral lesions diagnosed within the first weeks of life according to primary treatment, initial status (donor and recipient), and to the condition of the co-twin (dead or alive).

This hemodynamic theory was strengtiiened by the consistent association of severe anemia in single survivors irrespective of their initial status of donor or recipient with polycythemia in the first-dead twin bodi postnatally and in utero within hours following death [14,17-21]. Jou et al. [22] and Gembruch et al. [23] have even reported significant blood transfusion into the agonizing twin, prior to deadi, using color and pulsed Doppler examination.

The development of anemia in the surviving twin can be accurately monitored using maximum velocities in the middle cerebral artery considering a cutoff value of or greater than 1.5 multiples of the median (MoM). This can allow correction of anemia by intrauterine transfusion [6] in response to fetal exsanguination. Aldiough this may appear to be a life-saving procedure, its influence on the development of cerebral lesions in the survivor needs to be further investigated [20,21].

In TTTS, amnioreduction interacts significandy with feto-fetal hemodynamics with acute changes in fetal blood pressure in bodi donors and recipients [24]. In cases of incomplete laser surgery of the placenta, feto-fetal blood transfusion can also occur and cause severe anemia and polycythemia, respectively in the affected twin pairs [9]. Hemodynamic instability exposes the two fetuses to the risk of developing vascular disruptive lesions, particularly in the brain. The padiophysiology can be two-fold: lowflow and high-flow lesions. Bodi low-flow and highflow injuries can result from chronic or acute situations and they can equally affect the donor and the recipient twin (Figure 1).

Anatomical distribution and imaging of fetal cerebral lesions depend upon duration and type of injury as well as upon maturity of the brain at the time of the insult [25]. The fetal brain responds with increasing astrocytic reaction with maturation [25]. Porencephalic cysts without gliosis are therefore more likely to occur in immature brains aldiough septated cysts with irregular walls due to an intense astrocytic proliferation are more likely to develop when the injury occurred during the late second or early third trimester of pregnancy [25]. Low-flow insults occurring before 28 weeks of gestation may alter the neuronal population and interrupt neuronal migration [14,26,27]. This will lead to the development of periventricular leukomalacia, multicystic leukoencephalopathy, and hemorrhage in the subependymal germinal matrix, which can extend into the lateral ventricles and into the cerebral parenchyma. Hemorrhage of the germinal matrix may be associated with venous congestion in the adjacent white matter resulting into parenchymal venous hemorrhagic infarction. These different entities can be isolated or found in association. From 36 weeks onwards, subcortical leukomalacia [27] and ulegyria can also be observed. Closer to term, acute hemodynamic imbalance can also affect the basal ganglia or result in the development of lenticulostriate vasculopathy [28] (Figure 2).

Figure 1. Clinical situations associated with low-flow and high- flow cerebral lesions.

Figure 2. Anatomical location and patterns of cerebral lesions related to twin-to-twin transfusion syndrome in relation with timing of the insult.

Hypervolemia in the recipient can weaken the walls of immature vessels and cause hemorrhagic damage to the germinal matrix that can also affect the periventricular white matter. This can also occur in donor fetuses as a consequence of acute temporary high-flow particularly following amnioreduction. Hemorrhage can also occur within an ischemie lesion at the time of revascularization of immature vessels (Figures 2 and 3) as a mirror mechanism of that described above in the recipient. Thus, the same anatomical findings can be described in both donors and recipients. These patterns (Table II) are not specific and belong to a large spectrum of vascular disruptive cerebral lesions of which ITlS is one of many causes.

The risk of developing cerebral lesions in single survivors was significantly lower following laser treatment (5.45% vs. 35%, p = 0.0027, Table II). This can be explained by the protection of the survivor from exsanguination into its dead co-twin and placenta by the coagulation of the chorionic plate anastomoses. The presence of anemia or the development of cerebral lesions in the survivor may all be explained by incomplete coagulation of the inter-twin anastomoses. Incomplete coagulation can be suspected in at least 22% of the cases with IUFD of one twin and therefore in at least 8.3% of all procedures [29].

Figure 3. Sonographic aspects of the main types of cerebral lesions associated with twin-to-twin transfusion syndrome using transabdominal and transvaginal in utero ultrasound. (A, B) Ischemia of left thalamus and caudate nuclei on an anterior coronal and parasagittal planes at 21 weeks; (C) necrosis of the corpus callosum on a midsagittal plane at 27 weeks; (D) periventricular leukomalacia on a coronal posterior plane at 27 weeks; (E) subcortical leukomalacia on an anterior coronal plane at 27 weeks; (F) multicystic leukoencephalopathy with ex-vacuo ventriculomegaly on a parasagittal plane at 27 weeks; (G) bilateral periventricular leukomalacia (grade IA) on an anterior coronal plane at 27 weeks; (H) left temporo-parietal atrophy on an anterior coronal plane at 25 weeks; (I) microcephaly with diffuse enlarged subarachnoid spaces on a parasagittal plane at 25 weeks; (J) intraventricular hemorrhage (grade II) with heterogeneity of the right plexus choroid and of the ipsilatcral subependymal area on a posterior coronal plane at 30 weeks; (K) mild ventriculomegaly on a parasagittal plane at 18 weeks; (L) bilateral intraventricular hemorrhage with extension into the left periventricular parenchyma (grade III) on an anterior coronal plane at 27 weeks.

Ultrasound examination and MRI (Figure 4) are complementary techniques in depicting fetal brain lesions at an early stage. One case (number 23) had both normal antenatal cerebral US and MRI; however mild ventriculomegaly with cystic periventricular leukomalacia was diagnosed at 5 days of life. This condition can be explained in utero by a missed transient parenchymal hyperechogenicity that usually evolves towards mildly symptomatic periventricular leukomalacia.

Figure 4. MR images of the main cerebral lesions associated with twin-to-twin transfusion syndrome. (A, B): diffuse necrosis of the brain with multicystic leukoencephalopathy, microcephaly, and bilateral ventriculomegaly at 27 weeks on axial (A) and coronal (B) MR scans (T2-weighted HASTE sequences); (C, D): microcystic periventricular lesions corresponding to periventricular leukomalacia associated with microcephaly on parasagittal MR scans (T2-weighted HASTE sequences). Conclusions

In addition to therapy, screening for brain lesions is an important component of the management of ‘1″1″1’S. Ultrasound and MRI have both proven useful and complementary techniques. Fetal cerebral MRI proved reliable when performed from 28 weeks onwards. The interval between the insult and visualization of fetal cerebral damage by US or MRI is a critical factor in fetal surveillance. This was of 3-6 weeks following the death of one twin but could extend to more than 10 weeks. And it is more difficult to date the insult leading to brain damage when both twins are alive. The risk of developing cerebral lesions in single survivors is significantly lower following laser treatment. Transient ultrasound features in the absence of any MRI finding have a good prognosis as shown in up to 23% (6/26) of the cases reported here. Cohort studies should strive, in addition to long-term neurodevelopment follow-up, to correlate perinatal imaging with long-term development in survivors.

References

1. Bejar R, Vigliocco G, Gramajo H, Solana C, Benirschke K, Berry C, Coen R, Resnik R. Antenatal origin of neurologic damage in newborn infants. II. Multiple gestations. Am J Obstet Gynecol 1990;162: 1230-1236.

2. Bajoria R, Wigglesworth J, Fisk NM. Angioarchitecture of monochorionic placentas in relation to the twin-twin transfusion syndrome. Am J Obstet Gynecol 1995;172:856-863.

3. Sebire NJ, Souka A, Skentou H, Geerts L, Nicolaides KH. Early prediction of severe twin-to-twin transfusion syndrome. Hum Reprod 2000;15:2008-2010.

4. Fusi L, McParland P, Fisk N, Nicolini U1 Wigglesworth J. Acute twin-twin transfusion: Possible mechanism for brain damaged survivors after intrauterine death of a monochorionic twin. Obstet Gynecol 1991;78:517-520.

5. Senat MV, Deprest J, Boulvain M, Paupe A, Winer A, Ville Y. Endoscopie laser surgery versus serial amnioreduction for severe twin-to-twin transfusion syndrome. N Engl J Med 2004;351:134-144.

6. Man G, Roberts A, Detti L, Kovanci E, Stefos T, Bahado-Singh RO, Deter RL, Fisk NM. Perinatal morbidity and mortality rates in severe twin-twin transfusion syndrome: Results of the international amnioreduction registry. Am J Obstet Gynecol 2001;185:708-715.

7. Graef C, Ellenrieder B, Hecher K, Kackeloer BJ, Huber A, Bartmann P. Long-term neurodevelopmental outcome of 167 children after intrauterine laser treatment for severe twin-twin transfusion syndrome. Am J Obstet Gynecol 2006; 194: 303-308.

8. Senat MV, Loizeau S, Couderc S, Bernard JP, VuIe Y. The value of middle cerebral artery peak systolic velocity in the diagnosis of fetal anemia after intrauterine death of one monochorionic twin. Am J Obstet Gynecol 2003; 189: 1320- 1324.

9. Robyr R, Lewi L, Yamamoto M, Deprest J, Ville Y. Permanent feto-fetal transfusion from the recipient to the donor twin. A complication of laser surgery in twin-to-twin transfusion syndrome. AmJ Obstet Gynecol 2004;191:S163.

10. Volpe J.J. Intraventricular hemorrhage and brain injury in the premature infant. Neuropathology and pathogenesis. Clin Perinatol 1989;16:361-386.

11. Pierrat V, Duquennoy C, van Haastert IC, Ernst M, Guilley N, de Vries LS. Ultrasound diagnosis and neurodevelopmental outcome of localized and extensive cystic periventricular leucomalacia. Arch Dis Child Fetal Neonat 2001; 84:F151-156.

12. Benirschke K. Twin placenta in perinatal mortality. NY State J Med 1961;61:1499-1508.

13. Bendon RW, Siddiqi T. Acute twin-to-twin in utero transfusion. Ped Pathol 1989;9:591-598.

14. Larroche JC, Droulle P, Delezoide AL, Narcy F, Nessmann C. Brain damage in monozygous twins. Biol Neonat 1990; 57:261-278.

15. Denbow M, Fogliani R, KyIe P, Letsky E, Nicolini U, Fisk N. Hematological indices at fetal blood sampling in monochorionic pregnancies complicated by feto-fetal transfusion syndrome. Prenat Diagn 1998;18:941-946.

16. Wee LY, Taylor MJ, Vanderheyden T, Wimalasundera R, Gardiner HM1 Fisk NM. Reversal of twin-twin transfusion syndrome: Frequency, vascular anatomy, associated anomalies and outcome. Prenat Diagn 2004;24:104-110.

17. Fusi L, Gordon H. Twin pregnancy complicated by single intrauterine death. Problems and outcome with conservative management. BrJ Obstet Gynaecol 1990;97:51 1-516.

18. Okamura K1 Murotsuki J, Tanigawara S1 Uehara S, Yajima A. Funipuncture for evaluation of hematologie and coagulation indices in the surviving twin following co-twin’s death. Obstet Gynecol 1994;83:975-978.

19. Nicolini U, Pisoni MP, Cela E, Roberts A. Fetal blood sampling immethately before and within 24 hours death in monochorionic pregnancies complicated by single intrauterine death. Am J Obstet Gynecol 1998;179:800-803.

20. Tanawattanacharoen S, Taylor MJ, Letsky E, Cox PM, Cowan FM, Fisk NM. Intrauterine rescue transfusion in monochorionic multiple pregnancies with recent single intra-uterine death. Prenat Diagn 2001;21:274-278.

21. Senat MV, Bernard JP, Loizeau S, Ville Y. Management of single death in twin-to-twin transfusion syndrome: A role for fetal blood sampling. Ultrasound Obstet Gynecol 2002;20: 360-363.

22. Jou HJ, Ng KY, Teng RJ, Hsieh FJ. Doppler sonographic detection of reverse twin-twin transfusion after death of the donor. J Ultrasound Med 1993;5:307-309.

23. Gembruch U, Viski S, Baganery K, Berg C, Germer U. Twin reversal arterial perfusion sequence in twin-to-twin transfusion syndrome after the death of the donor co-twin in the second trimester. Ultrasound Obstet Gynecol 2001; 17: 153-156.

24. Smith JF, Pesterfield W, Day LD, Jones RO. Doppler evidence of improved fetoplacental hemodynamics following amnioreduction in the stuck twin phenomenon. Obstet Gynecol 1997;90:681-682.

25. Barkovich AJ. Brain and spine injuries in infancy and childhood. Congenital malformations of the brain and the skull in pethatrie neuroimaging. Chapters IV and V. In: Barkovich AJ, editor. Pethatrie neuroimaging. Third ed. Philadelphia: Lippincott, Williams & Wilkins; 2000. pp 157-249.

26. Larroche JC, Girard N, Narcy F, Fallet C. Abnormal cortical plate (polymicrogyria), heterotopias and brain damage in monozygous twins. Biol Neonate 1994;65:343-352.

27. Levene MJ, Chervenak FA, Whittle M, Benett MJ, Punt J, editors. Fetal and Neonatal Neurology and Neurosurgery. 3rd ed. Edinburgh, Scotland: Churchill Livingstone; 2 l:pp 323-404.

28. de Vries LS, Beek FJ, Stoutenbeek P. Lenticulostriate vasculopathy in twin-to-twin transfusion syndrome: Sonographic and CT findings. Pediatr Radiol 1995;25:S41-42.

29. Cavicchioni O, Yamamoto M, Robyr R, Takahashi Y, Ville Y. Intrauterine fetal demise following laser treatment in twin-to-twin transfusion syndrome. BJOG 2006; 11 3: 590-594.

EDWIN QUARELLO1, MARC MOLHO2, & YVES VILLE1

1 Department of Obstetrics and Gynecology, CHI Poissy St Germain- en-Laye, Universite Paris-Ouest, Poissy, France and

2 Department of Radiology, CHI Poissy St Germain-en-Laye, Universite Paris-Ouest, Poissy, France

(Received 26 June 2006; revised 1 February 2007; accepted 1 1 April 2007)

Correspondence: Pr Y. Ville, Service de Gynecologie Obstetrique, Hopital de Poissy, 10 Rue du Champ Gaillard, 78303 Poissy, France.

E-mail: [email protected]

Copyright Taylor & Francis Ltd. Aug 2007

(c) 2007 Journal of Maternal – Fetal & Neonatal Medicine. Provided by ProQuest Information and Learning. All rights Reserved.