Before 14 weeks' gestation, separation between amnion and chorion is normal. Those two membranes usually fuse between 14 and 16 weeks' gestation, and the anechoic space between them disappears. Lack of fusion of amnion and chorion has been reported previously, particularly after amniocentesis.1–3 After 16 weeks' gestation, persistent separation of the membranes is rare and has been associated with preterm delivery, fetal death, and recently, chromosomal abnormalities.1–6 Some reports described extra-amniotic pregnancies in which the amniotic membrane forms a band and collapses centrally around the base of the umbilical cord.3,4 In this study, our objective was to evaluate sonographic findings, clinical courses, and obstetric outcomes in 15 cases of sonographically detected amnion-chorion separation that continued after 17 weeks' gestation.
Materials and Methods
We searched our ultrasound database for information on all pregnant women (unselected population) with live fetuses and complete separation between amnion and chorion that persisted beyond 17 weeks' gestation. Complete separation was defined as the amnion being separated from the chorion on at least three sides of the gestational sac, with an anechoic space between them (Figure 1). This condition appears very different from a subchorionic hematoma, in which the membrane is thick and the hematoma causes a mass effect on the outside of the sac. From December 1991 to December 1998, 15 women had complete amnion-chorion separation seen sonographically at or after 17 weeks. Medical records were reviewed for whether women had amniocenteses, the results of those who did, and the outcomes of the pregnancies.
Scans were done with an Acuson 128XP-10 (Acuson, Mountain View, CA), with a 3.5- or 5-MHz transducer, and a Sequoia Acuson system (Acuson, Mountain View, CA). During that time, approximately 89,700 obstetric scans were done at our facility.
Study patients were referred for prenatal sonography for many indications, including two for advanced maternal age, two for second opinions on floating membranes, four for fetal growth assessment, six for increased risk of or suspected congenital defects, and one for decreased fetal activity. Gestational age at diagnosis of amnion-chorion separation was 17–37.5 weeks. Ten women were diagnosed before 24 weeks gestation and five at greater than 29 weeks. Two women did not have amniocenteses during gestation. Two women had amniocenteses after identification of amnion-chorion separation. Another woman had an attempted amniocentesis after the identification of amnion-chorion separation, but it was unsuccessful in obtaining amniotic fluid. The remaining ten women had amniocenteses before identification of amnion-chorion separation. Three fetuses had trisomy 21, all of whom had additional fetal abnormalities on scans. One of those fetuses was evaluated in the third trimester and had mild pyelectasis, dilatation of the small bowel, and polyhydramnios, which were complicated by preterm labor and delivery at 32 weeks' gestation. Of the two second-trimester fetuses with trisomy 21, one had a cystic hygroma and heart defect, whereas the other had a nuchal fold, pyelectasis, and heart defect. One of the second-trimester fetuses with Down syndrome was aborted and the other was lost to follow-up. Two of three fetuses with trisomy 21 (17 weeks' and 32 weeks' gestation) had amnion-chorion separation before amniocenteses were done. Nine fetuses had normal karyotypes and three others were not suspected of aneuploidy at delivery.
Two women had several normal results of scans after unremarkable amniocenteses in which no separation of amnion from chorion was identified. Amnion-chorion separations only became apparent late in the third trimester in those cases.
Three fetuses died between 24 and 25 weeks' gestation. One of those had a two-vessel cord. Although the three women had second-trimester amniocenteses (including one in which amniocentesis was not successful after amnion-chorion separation had been identified), the fetuses died 6–8 weeks after amniocentesis.
Several different adverse perinatal events occurred. Two fetuses had growth restriction (estimated fetal weight no more than tenth percentile). There were five preterm deliveries. Two women had preterm labor and delivery at 32 and 36 weeks' gestation, and a third had preterm rupture of membranes followed by delivery at 35 weeks' gestation. One woman had a cesarean at 29 weeks for growth restriction, oligohydramnios, and abnormal Doppler waveform tracing. Another delivered at 34 weeks for worsening oligohydramnios. Five infants were delivered at term and are alive and well, although one was delivered at 37 weeks for fetal cardiac decelerations and abruption. In total, ten of 15 pregnancies with amnion-chorion separation resulted in live newborns, one of whom has Down syndrome.
Lack of fusion of amnion and chorion is seen sonographically as a thin membrane (amnion) floating freely in the amniotic fluid, with varying degrees of collapse onto the fetus. In some cases, it leads to amniotic band syndrome or an extra-amniotic pregnancy in which the fetus penetrates the chorionic space, leaving the amnion freely floating at the base of the umbilical cord. Ashkenazy et al3 found that it can occlude the umbilical cord secondary to amniotic band formation. Graf et al4 found amniotic bands forming from membrane separation after fetal surgery, compromising umbilical cord blood flow and causing fetal death. That might be how three of the fetuses in the present study died at 24–25 weeks. Bronshtein and Zimmer5 further reported oligohydramnios, which is present occasionally with amnion-chorion separation because some of the amniotic fluid is outside the amniotic membrane. Fetal malformations were observed in all nine of their cases.
Amnion-chorion separation has also been associated with chromosomal anomalies such as Down syndrome and cystic hygromata.5,6 Appelman et al (Appelman Z, Zalel Y, Fried S, Caspi B. Delayed fusion of amnion and chorion: A possible association with trisomy 21 [letter]. Ultrasound Obstet Gynecol 1998;11:303–5.) reported five fetuses with amnion-chorion separation, four of whom had Down syndrome. Among 13 fetuses with amnion-chorion separation, Kaufman et al6 found one with a large cystic hygroma and presumed chromosomal abnormality.
We previously presented a preliminary report on five women with amnion-chorion separation.1 When the present study of 15 subjects is combined with our previous consecutive study, we have the largest reported single-institution experience to date of this rare membrane abnormality. Three of 20 fetuses (15%) had Down syndrome. Four of 20 women lost their karyotypically normal fetuses between 21 and 25 weeks' gestation. Seven of 20 women (35%) delivered prematurely, and of the fetuses, two had fetal growth restriction and two had premature rupture of membranes. Thirteen of 20 women had amniocenteses predating identification of amnion-chorion separation and seven (35%) did not. Two of the seven women without amniocentesis before identification of complete amnion-chorion separation had fetuses with Down syndrome. Only seven of 20 (35%) delivered at term.
Levine et al2 recently reported seven fetuses with amnion-chorion separations, resulting in three fetal deaths, two preterm deliveries at 29 and 35 weeks, and two term deliveries. Their study focused specifically on the relationship of this abnormality to amniocentesis, because all seven subjects had procedures for karyotypic analysis. The finding that 33% of subjects in the current study (35% when both our studies are combined) did not have amniocenteses before the initial observations of amnion-chorion separation suggests that membrane separation is not necessarily a complication of amniocentesis. Because of the high incidence of Down syndrome in our study and in others, amniocentesis should be considered for cases of spontaneous amnion-chorion separation that persist beyond 17 weeks. Amniocentesis is technically more difficult to do when the amnion is not fused to the chorion, but karyotypic analysis might be indicated in those women. Further studies are necessary to determine whether there is a greater risk of fetal death when amniocentesis is done after amnion-chorion separation, compared with fused membranes.
We found that complete separation of amnion from chorion is rare, associated with fetal death, preterm labor, preterm delivery, premature rupture of membranes, and fetal growth restriction. Fewer than half our subjects delivered normal newborns at term. Also alarming was the high incidence of trisomy 21 in the fetuses, particularly among those in whom ultrasound findings preceded amniocentesis. Further experience is needed to determine whether amnion-chorion separation should be used (as an isolated finding) as an indication for karyotyping.
1. Benacerraf BR, Frigoletto FD Jr. Sonographic observation of amniotic rupture without amniotic band syndrome. J Ultrasound Med 1992;11:109–11.
2. Levine D, Callen PW, Pender SG, McArdle CR, Messina L, Shekhar A, et al. Chorioamniotic separation after second-trimester genetic amniocentesis: Importance and frequency. Radiology 1998;209:175–81.
3. Ashkenazy M, Borenstein R, Katz Z, Segal M. Constriction of the umbilical cord by an amniotic band after midtrimester amniocentesis. Acta Obstet Gynecol Scand 1982;61:89–91.
4. Graf JL, Bealer JF, Gibbs DL, Adzick HS, Harrison MR. Chorioamniotic membrane separation: A potentially lethal finding. Fetal Diagn Ther 1997;12:81–4.
5. Bronshtein M, Zimmer EZ. Oligohydramnios with amnio-chorionic separation at 15–16 weeks' gestation. Prenat Diagn 1995;15:161–4.
6. Kaufman AJ, Fleischer AC, Thieme GA, Shah DM, James AE Jr. Separated chorioamnion and elevated chorion: Sonographic features and clinical significance. J Ultrasound Med 1985;4:119–25.