1991-05-06-11 Twin, acardiac, ultrasound-guided embolization © Roberts www.thefetus.net/
Twin, acardiac, ultrasound-guided embolization
Richard M. Roberts, PhD, MD*, Dinesh M. Shah, MD,Philippe Jeanty, MD, PhD, James F. Beattie Jr., MD
Address correspondence to:Richard M. Roberts, PhD, MD, Department of Pediatrics, University of Tennessee Medical School, Chattanooga Unit, 910 Blackford St., Chattanooga, TN, 37403-1299. Ph 615-778-6112 Fax 615-778-6215; ¶ University of Texas, San Antonio, Dept. of OB/GYN; § Vanderbilt University Dept. of Radiology, Nashville, TN; ¤Hutcheson Medical Center, Ft. Oglethorpe, Georgia
Synonyms: Acardiac monster (pejorative); TRAP sequence; acardiac parabiotic twin; acardiac anomaly; holoacardius, hemicardius; chorioangiopagus parasiticus; acardiac acephalic twin; acardius anceps, acardius acephalus, acardius acormus, acardius amorphus; fetus amorphus anideus; acardia mylacephalus.
Prevalence: 0.3:10,000 deliveries1, estimated to occur in 1% of monochorionic twin births2, and three times more likely to occur among monozygous pairs of triplet births3. In 1988, 86,315 twin births and 2,366 other multiple births occured in the U.S.; estimating that one-third of these are monozygous (MZ), and two-thirds of MZ twins are monochorionic, 196 TRAP sequences per year is a rough estimate23.
Definition: Monochorionic twin, triplet or other multiple pregnancy (mono- or diamnionic) characterized by a pump sibling perfusing an anomalous recipient sibling via an artery-to-artery anastomosis, with reversed direction of flow of arterial blood to the anomalous fetus, which is either acardiac or has severely anomalous cardiac structure.
Etiology: The prominent theory holds that artery-to-artery anastomosis leads to reversed perfusion of one twin with resultant disruption of development of the cardiovascular system and a cascade of disruption of organ development in the perfused twin. However, the anomalous twin may have an underlying cytogenetic anomaly, and a primary malformed cardiovascular system in the perfused twin prior to anastomosis may be etiologic in some instances.
Associated anomalies: See text.
Differential diagnosis: Monozygous twin-to-twin transfusion syndrome; single twin demise of different etiology.
Prognosis: Dependent on whether signs of decompensation in the pump twin occur, and how early in gestation adverse effects such as polyhydramnios occur. Once polyhydramnios develops, the risk of ruptured membranes and premature delivery is high; sudden fetal demise of the normal twin may occur even in the absence of cardiac decompensation; a significant proportion do not survive—perhaps higher than 50%. Recent evidence suggests that the prognosis worsens in direct relation with the relative size of the anomalous twin4.
Recurrence risk: Not significantly increased.
Management: Options include 1) no intervention, 2) termination of pregnancy, 3) serial ultrasounds to monitor for signs of decompensation, 4) medical management of polyhydramnios or management via serial amniocenteses, 5) digitalization for prophylaxis or for treatment of cardiac failure, 6) hysterotomy for removal of the anomalous twin, 7) endoscopic clamping of the anomalous twin"s cord (not yet attempted), 8) laser coagulation of the anastomosis if placental (not yet attempted), and 9) embolization of the circulation of the anomalous twin. Attempted feticide of an anomalous hemicardiac twin by cardiac puncture has proven futile and would present a danger to the normal twin5; any substance injected into the acardiac circulation could circulate to the normal twin. Massive anomalous twins have caused dystocia6 and uterine rupture7.
MESH Diseases-in-Twin-diagnosis; Fetofetal-Transfusion-diagnosis; Heart-Defect, Congenital-diagnosis; Pregnan- cy, -Multiple; Monsters; Umbilical cord blood supply ICD9 761.5 CDC 758.456
Acardiac twins were first documented in 15338. Artery-to-artery anastomosis occurs in approximately 1% of monochorionic twin pregnancies (and a higher percentage of triplet pregnancies), leading to reversed perfusion of one twin with resultant disruption of development of the organ systems of the perfused twin—hence the mnemonic TRAP (Twin Reversed Arterial Perfusion) sequence9. The reversed perfusion was first described in 185016 and confirmed by Doppler ultrasonography in 198811. An acardiac twin was first diagnosed with ultrasound in 197812. Potential intervention by embolization of the perfused twin was first suggested in 198311, and successfully performed with coil embolization in 1989 (steel coil) and 1990 (platinum coil)18,19. Hysterotomy with removal of the anomalous twin was first reported in 198920.
A-21-year-old G1P0 female (spouse non-consanguineous) was referred at 23 weeks gestation after a 2-week history of polyhydramnios, with a level I ultrasound diagnosis of fetal demise of a highly hydropic anomalous twin. The ultrasound revealed male twins, apparently monoamnionic monochorionic. One twin appeared normal, while color flow Doppler demonstrated reversed umbilical artery flow to an acardiac acephalic twin. Other initial findings included an allantoic duct remnant, at first thought to represent a third umbilical vessel, but correctly recognized after Doppler revealed no flow inside. A small cyst was also seen at the insertion of the cord. The two umbilical cords had a common placental insertion. Membranes originated from the two umbilical cords, and another membrane appeared attached between the lower extremities of the anomalous twin, suggesting amnionic sheets or bands. The anomalous twin had no upper extremities or cranium. The acardiac hydropic thorax was massively enlarged; with a huge posterior cystic mass, larger than the dysplastic rib cage, and a lateral subcutaneous cystic mass bilaterally. The number of vertebrae was reduced. A single short ossified structure could represent a rudimentary right humerus or scapula. The bladder was abnormally small. The edematous lower extremities expanded and flexed one foot presented a severe equinovarus malformation. The genitalia were normal for a male.
Figure 1: Sagittal view of the acardiac fetus. Note the absence of head, and the large dorsal cyst.
Figure 2: Three "vessels" are seen at the insertion of the cord. One represents an allantoic duct remnant. Note the allantoic cyst (AC). UV = umbilical vein.
The patient was placed on indomethacin for medical management of the polyhydramnios. She was counseled that the risks of sudden demise or premature delivery in TRAP sequence with symptoms of decompensation in the second trimester were great, even with control of the polyhydramnios and digitalization to prevent or treat congestive heart failure. The following alternatives were presented to the family, with explanation of potential risks and benefits:
- Intervention limited to management of polyhydramnios (indomethacin and possible serial therapeutic amniocenteses), and prophylactic digitalization to try to prevent cardiac failure in the normal twin.
- Elective termination of both twins
- Hysterotomy and removal of the abnormal twin
- Endoscopic clamping of vessels to the abnormal twin under ultrasonic guidance
- Insertion of a platinum coil22 into the abnormal twin"s umbilical artery to block flow.
Risks of no intervention or of medical management only: demise likely before viability, or premature delivery with potential sequelae secondary to complications of extreme prematurity (including brain damage, possible blindness, deafness, mental retardation).
Hysterotomy risks: complications of major surgery; premature delivery and associated risks of prematurity; undefined risks of mid-trimester hysterotomy with regard to future fertility and survival of the normal twin.
Endoscopic cord clamping: premature delivery (probably less risk than that of hysterotomy) with risks of complications secondary to prematurity.
Platinum coil embolization of blood flow to the abnormal twin: possible risk of twin embolization syndrome damaging the normal twin, perhaps resulting in mental retardation.
Figure 3: Color Doppler of the common insertion of the cords. Note that the cord of the acardiac fetus (lwhite arrow) is smaller than the normal cord.
Figure 4: Post mortem view that demonstrates the coil into the umbilical artery of the acardiac twin.
Figure 5: The allantoic cyst at the root of the cord. The coil is also visible.
|Figure 6: The membranes that partially enveloped the origin of the cord and the limbs of the acardiac fetus.|
|Figure 7: The first coil has been placed (black arrow) and the flow stops proximal to the coil.|
The patient elected to have the embolization. An ultrasound-guided embolization procedure was performed, during which 4 platinum coils were inserted into the umbilical artery of the anomalous twin utilizing a 20 gauge needle. Confirmation of correct needle placement was obtained by injection of normal saline and observing retrograde flow into the fetus. The anomalous umbilical vein descended caudad adjacent to the umbilical artery rather than the normal cephalad and dorsal path towards the ductus venosus. Continuous color-flow Doppler monitoring indicated immediate and complete cessation of blood flow to the acardiac twin as soon as the first coil was placed.
The patient was discharged the next day, after ultrasound confirmed the absence of cord flow to the anomalous twin. The other twin had normal activity and appearance. During the two days post-procedure, she reported a total weight loss of 10 lbs. The second day she no longer felt fetal movements and returned for ultrasound examination, which revealed demise of the normal twin. The absence of signs of postmortem skin disruption or subcutaneous edema indicated a demise within the last 24 hours. No ultrasonographic cause of demise was detected. The following day, the patient elected to have labor induced by her primary care physician, and delivery occurred after approximately 12 hours.
Figure 8: An aspiration of the dorsal cyst is made, in order to karyotype non-hematological tissue.
Figure 9: Macroscopic appearance of the two twins.
The autopsy and dysmorphologic evaluation revealed no major or minor malformations, signs of congestive heart failure, emboli, or disseminated intravascular coagulation in the pump twin, and radiographs were normal. The arterial anastomosis was noted just within the placenta at the base of the umbilical cords. The acardiac acephalic twin was affected with the anomalies defined by ultrasound evaluation, with the following two exceptions: the amniotic band thought to exist between the lower extremities of the anomalous twin was actually one of several sheets of amnionic membrane attached to the umbilical cords near the placental insertion, which we interpret as incomplete development of two amnionic cavities. The anomalous twin also had complete syndactyly of the left 4th and 5th toes. The absence of lungs, heart, liver and spleen were confirmed.
Three platinum coils were adjacent to the umbilical artery, and one was partially outside of the vessel. No evidence of significant hemorrhage was present. The abdominal aorta was absent. The anomalous umbilical artery branched into many vessels after turning cephalad. A urachus with protrusion at the umbilical ring was present. Karyotype of cord blood was normal, Cell culture of fluid aspirated from the cystic hygroma at the time of embolization and from fascia postmortem from the anomalous twin was unsuccessful.
Two distinct etiologic mechanisms have been proposed since the 19th century: 1) both twins may initially be normal, and the anastomosis enables the heart of the twin which is more advanced in development to disrupt normal cardiac development of the recipient twin; 2) the anomalous twin initially has abnormal cardiac development, either from genetic or non-genetic etiology, which is further compromised and finally overcome by abnormal pressures from retrograde flow from the pump twin via the artery-to-artery anastomosis.
Because both karyotypically normal twin pairs as well as discordant twin pairs (the acardiac twin always being cytogenetically abnormal) have been documented, it is possible that the condition is, in fact, etiologically heterogeneous and both theories are correct.
Cardiac anomalies are common in cytogenetically anomalous fetuses as well as in cytogenetically normal monozygous twins. Anomalous development of the cardiovascular systems in some of these twins—in the present case the umbilical vein descended caudad from the umbilical insertion and the abdominal aorta was absent—suggests that cardiovascular anomalies are not solely the result of reversed perfusion. However, the pattern of malformation of these twins is clearly heavily influenced by the reversed perfusion. The anomalous twin receives blood from an artery-to-artery anastomosis and hence receives oxygen- and nutrient-poor blood which has already been circulated through the pump twin. Since the richest blood arrives through the iliac arteries, there is a relative sparing of the lower extremities and anomalies related to anoxia, and poor nutrient supply increases with increased distance from the entry of the vessels. Blood returns from the anomalous twin via its umbilical vein and anastomoses either with the umbilical vein of the normal twin or with venous vessels in the placenta.
Claudius pointed out in 1859 that the allantoic stalk of twin gestations reaches the chorion at different times, and anastomosis between allantoic arteries could occur then. Early fusion could result in complete disruption of the normal cardiovascular system prior to development of the cardiac flexure, and later fusion could disrupt an already-present primitive heart. Cytogenetically anomalous twins could grow at a slower rate, and hence would be succeptible to "takeover" via the arterial anastomosis, accounting for the significant proportion of karyotypically anomalous twins. In cytogenetically and phenotypically normal twin embryos, whichever twin"s heart begins effective circulation first, with greater pressure, would result in retrograde flow to the cardiovascularly more immature twin. Similarly, lower blood pressure could result from a unilateral cardiac defect, with onset of retrograde flow from the normal twin.
Essentially every single organ system is affected, but the following is a list of the most characteristic anomalies. The heart may be completely absent, reduced to a tubular or even a 2-chamber heart. A single umbilical artery is the rule. The lower extremities of the abnormal twin are usually spared the most, presumably because the femoral artery contains relatively the most oxygen and nutrients. Malformations of the lower extremities include ray deficiencies and equinovarus malformation. The abdominal visceral anomalies include absence of any discernable abdominal organs, persistence of embryonic structures like the urachus, and aberrant vasculature. The external genitalia may be normal, deficient or absent. The thoracic structures, including ribs and vertebrae, may be present, rudimentary, severely dysplastic or absent. Marked edema and cystic hygromata which may attain massive proportions are typical. Acardiac fetuses typically are acephalic with absent upper extremities. When the head is present, many malformations have been described, including anencephaly, holoprosencephaly, and facial clefting.
The normal (pump) twin may develop cardiac failure with polyhydramnios, cardiomegaly, pericardial, pleural, and peritoneal effusion, engorged umbilical vein, and subcutaneous edema. Heart failure may occur in the absence of polyhydramnios, however. The pump twin is also at risk of those malformations associated with monozygous twinning4, and of disseminated arterial calcification (reported once)5.
Affected pregnancies are at risk of premature delivery secondary to polyhydramnios and of sudden demise of the apparently normal pump twin, which may occur even in the absence of signs of cardiac decompensation.
A significant percentage of anomalous twins in TRAP sequence are chromosomally anomalous and may not be genotypically identical; the embryologic origin remains controversial. The first cytogenetic discordant twins were reported in 196619.
One remarkable case provides firm evidence that polar-body fertilization occurs in man, by documenting a triploid acardiac twin, the result of fertilization of a single ovum by two genetically dissimilar sperm, one uniting with the first mitotic polar body20.
At the time of the procedure, we were aware of one other embolization procedure for this indication which had been done two months prior, with growth of the anomalous twin ceasing and continued normal growth of the normal twin post-procedure. Subsequently, this twin successfully went to term, as occurred in another case report from the Japanese literature16.
It is clear that demise of the pump twin can occur post-procedure even in the absence of any signs of cardiac failure21; in our case, the normal fetus had no ultrasonic signs of failure—no effusion, cardiomegaly, or engorgement of the umbilical vein or inferior vena cava, and no signs whatsoever of emboli or disseminated intravascular coagulation at autopsy. We theorized that the sudden elimination of the parasitic circulation after embolization would relieve strain on the normal heart, and given that the patient lost 10 lbs in the ensuing two days suggests resolving polyhydramnios, hence relief of cardiovascular strain—yet the normal twin died.
Retrospectively, it might have been prudent to digitalize prior to this procedure in hopes of reducing any unforseen potential shock to the cardiovascular system of the normal twin. However, unexplained demise of a monozygous twin occurs more commonly than in singleton pregnancies (as may be the case in instances of fetus papyraceous), and it is possible that the demise was not related to the procedure.
Because of their rarity and the absence of a registry on prenatal management of acardiac twins, several management questions are difficult to answer. What is the relative risk of demise or premature delivery when polyhydramnios first occurs—is it markedly increased in 2nd trimester pregnancies versus 3rd trimester pregnancies? Are the benefits of an embolization procedure identical to those derived from hysterotomy? Are the risks of fetal demise post-embolization the same as post-hysterotomy? Is hysterotomy preferable, removing a significant possibility of transfer of toxic products or emboli to the normal twin, or does embolization prevent any potential circulation of toxic products or emboli as efficiently as removal with hysterotomy? Because twin embolization phenomena have not been reported in acardiac twinning, such risk is not similar to that in twin to twin transfusion syndrome and is undefined post-hysterotomy or post-embolization procedure. The risk of emboli thrown post-embolization procedure or post-hysterotomy/removal may be very low, and can only be defined with a registry of these procedures. We would be pleased to act as a registry for therapy in acardiac twin pregnancy.
Acknowledgements: We are grateful to Dr. Karim Iskander for referral of his patient and to Dr. Richard Porreco for sharing information on his embolization procedure.
1. Gillim DL, Hendricks CH. Holoacardius: review of the literature and case report. Obstet Gynecol 2:647-653,1953
2. Napolitani F, Schreiber D, Schreiber I: The acardiac monster. Am J Obstet Gynecol 80:582, 1960.
3. James W. A note on the epidemiology of acardiac monsters. Teratology 16:211-216, 1977.
4. Schinzel A, Smith DW, Miller J: Monozygotic twinning and structural defects. J Pediatr 1979; 95:921-930.
5. Royston D and Geoghegan F. Disseminated arterial calcification associated with acardius acephalus. Arch Dis Child 58:641-43, 1983.
6. Moore TR, Gale S, Benirschke K: Perinatal outcome of forty-nine pregnancies complicated by acardiac twinning. Am J Obstet Gynecol 163:907-912, 1990.
7. Seeds JW, Herbert WNP, Richards DS. Prenatal sonographic diagnosis and management of a twin pregnancy with placenta previa and hemicardia. Am J Perinatol 4:313-316, 1987.
8. Loughead Jr. JR, Halbert DR. An acardiac amorphous twin presenting soft tissue dystocia. South Med J 62:1140-1142, 1969.
9. Boronow RC, West RH. Monster acardius parasiticus. Am J Obstet Gynecol 88:233, 1964
10. Benedetti: De dingulis corpori humani morbis a capite ad pedis. 26:407, 1533.
11. Van Allen MI, Smith DW, Shepard TH. Twin reversed arterial perfusion (TRAP) sequence: a study of 14 twin pregnancies affected with acardius. Semin Perinatol 7:285-293, 1983
12. Hempel D. De monstris acephalis. Dissert and Hafniae, 1850. Cited by Warkany J. Congenital malformations: Notes and comments. Year Book Medical Publishers, Chicago, p 473, 1971
13. Pretorius DH, Leopold GR, Moore TR, et al. Acardiac twin. Report of doppler sonography. J Ultrasound Med 7:413-416, 1988.
14. Lehr C, DiRe J: Rare occurrence of a holoacardious acephalic monster: sonographic and pathologic findings. JCU 6:259-261,1978.
15. Platt LD, DeVore GR, Bieniarz A et al. Antenatal diagnosis of acephalus acardia: a proposed management scheme. Am J Obstet Gynecol 146:857, 1983
16. Hamada H, Okane M, Koresawa M et al. [Fetal therapy in utero by blockage of the umbilical blood flow of acardiac monster in twin pregnancy.] Nippon Sanka Fujinka Gakkai Zasshi 41:1803-1809, 1989.
17. Porreco RP, Barton SM, Havercamp AD. Occlusion of umbilical artery in acardiac, acephalic twin. Lancet i:326-327, 1991.
18. Robie GF, Payne GG, Morgan MA. Selective delivery of an acardiac, acephalic twin. N Engl J Med 320:512-513, 1989.
19. Kerr M, Rashad M: Autosomal trisomy in a discordant monozygotic twin. Nature 212:726, 1966.
20. Bieber FR, Nance WE, Morton CC et al. Genetic studies of an acardiac monster: evidence of polar body twinning in man. Science 213:775-777, 1981
21. Fusi L, Fisk N, Talbert D et al: When does death occur in an acardiac twin? J Perinat Med 18:223-7, 1990.
22. Complex Helical Platinium Coil, Vascular Occlusion System Coil. Product number 311024, Kit product number: 520024. Target Therapeutics, 130 Rio Robles, San Jose, CA 95134-1806. Ph: 408-435-7000 / 800-345-2498; Fax: 408-943-0846.
23. The National Center for Health Statistics. Vital Statistics of the U.S., Natality, Volume 1:181,1988.