INTERNATIONAL INSTITUTE FOR THE TREATMENT OF
TWIN-TO-TWIN TRANSFUSION SYNDROME
"..so that babies and families who are suffering today will live and be happy"
Julian E. De Lia, MD, FACOG, FACS
The International Institute for the Treatment of Twin-to-Twin Transfusion Syndrome
Twin-To-Twin Transfusion Syndrome:
Pathohysiology and Placental Laser Therapy
Julian E. De Lia, M.D.
Associate Clinical Professor
Medical College of Wisconsin
12-13 September 2002
With the advent of ultrasound scanning, obstetricians can now detect and treat abnormalities during early pregnancy that may threaten continued intrauterine survival of the fetus, or its survival immediately postpartum. Some of these are abnormalities of the placenta, with the normal fetus threatened secondarily (see De Lia JE: Surgery of the placenta and umbilical cord. Clin Obstet Gynecol 1996;39:607-25). In order to undertake placental therapy it is essential to have a sound foundation in normal placental anatomy and, more importantly, an understanding of the pathologic findings therein. However, in some centers the ultrasound has led to a paradigm shift, where these disorders are perceived as abnormal video display images rather than maternal-fetal-placental pathologic processes.
Fetoscopically delivered laser energy is now available to treat placental abnormalities such as symptomatic placental chorioangiomas, and the consequences of the chorioangiopagous status of monochorionic twins: twin reversed arterial perfusion (acardius syndrome or TRAP) with heart failure in the 'pump' twin (see Arias F, et al.: Treatment of acardiac twinning. Obstet Gynecol 1998;91:818-21), and the twin-twin transfusion syndrome (see De Lia J, et al.: Fetoscopic laser occlusion of chorioangiopagus in severe twin transfusion syndrome. Obstet Gynecol 1990;75:1046-53). To date, laser surgery has been used predominantly in twin-twin transfusion syndrome (TTTS), therefore this presentation will focus on TTTS pathophysiology and the role of fetoscopic laser occlusion of chorioangiopagous vessels (FLOC procedure) in its treatment.
Syndrome in Previable Gestations
Monozygotic (MZ) or identical twin pregnancies experience high morbidity and mortality rates, particularly if the twinning process occurs more than 4 days from fertilization. Diamnionic monochorionic (DiMo) twins result from 4- to 8- day twinning and are the most common of MZ twins. Here, the twins share a single or monochorionic (MC) placenta, which explains excess morbidity and mortality when MZ multiples are compared to fraternal (dizygous) twins or MZ twins with separate (dichorionic) placentas. Since the identical twinning process occurs within days of conception and has no known etiology or cause, opportunities for physicians or couples to exercise primary prevention of MZ twinning complications are absent.
Although the majority of MC twin pregnancies are normal, approximately 20% suffer complications that can be traced to placental factors. The two MC placental factors include abnormal blood vessels that connect the twins in the placenta (chorioangiopagous vessels), and the tendency of their single placenta (chorion surface and villi) to be shared either equally or asymmetrically by the twins (see Machin G, et al.: Correlations of placental vascular anatomy and clinical outcomes in 69 monochorionic twin pregnancies Am J Med Genet 1996;61:229-36). The chorioangiopagous vessels, which seem to occur by chance, can lead to the classically described findings of chronic TTTS from imbalanced blood flow between the twins. Severe chorion asymmetry may lead to growth restriction and possible death of the twin with a small placental share. The asymmetry may be caused by the variations in twin MC embryoblasts’ orientation at implantation (see Gaziano E, et al.: Diamnionic monochorionic twin gestations: An overview. J Mat-Fetal Med 2000;9:89-96). If one MC twin dies, a sudden hemorrhage (acute transfusion) may occur from the live co-twin through the chorioangiopagous vessels to the dead twin. Death or significant injury (ischemic necrosis of somatic structures) from hypovolemia in the surviving twin may then follow (see Okamura K, et al.: Funipuncture for evaluation of hematologic and coagulation indicies in the surviving twin following co-twin's death. Obstet Gynecol 1994;83:975-8).
The random variations in vascular anastomoses and placental symmetry produce a syndrome that appears predominantly in previable pregnancies (but may also occur at later gestational ages with even greater clinical consequences), with varying degrees of growth discordance and amniotic fluid changes that may fluctuate in severity and, rarely, resolve spontaneously. Monochorionic twin pregnancies with significant placental asymmetry may explain some cases in which ultrasonographic and placental findings are consistent with TTTS, but have paradoxical or normal hematocrit values at birth or with in utero umbilical cord blood sampling (see Mari G, et al: "Pseudo" twin-twin transfusion syndrome and fetal outcome. J Perinatol 1998;18:399-403). It is not currently possible to determine by ultrasound the relative contribution of transfused blood versus placental asymmetry in TTTS cases, but the morbid fetal effects are similar and ultimately depend on the chorioangiopagous status of the twins.
The number of MC twin pregnancies at risk of TTTS in the United States (using CDC statistics) is approximately 3,800, since 1:34 births (2.93%) are plural based on 2000 birth record analysis (118,900 twin births X .33 identical X .66 monochorionic X .15 TTTS incidence). Recall that pregnancy terminations and spontaneous losses prior to 20 weeks go uncounted, therefore this estimate may be conservative. Understandably, infertility treatments have contributed to the rate of multiple birth. However, they have not reduced the incidence of identical twins even though multiple embryos are often produced. In contrast, studies show a higher rate (3 – 15 X) of identical twins than would occur in natural pregnancy in patients having these treatments (see Derom C, et al.: Increased monozygotic twinning rate after ovulation induction. Lancet 1987;1:1236-8, and Behr B, et al.: Blastocyst-ET and monozygotic twinning. J Assisted Reprod Genet 2000;17:349-51)
Patients with MC placental abnormalities present most often with a 'large-for-dates' uterus, and subsequently show ultrasound evidence of MC twins: like-sex fetuses, a single placental disc, and a thin (amnions only) interfetal membrane septum. Findings indicative of TTTS and/or placental asymmetry include: discordant fetal growth, discordant amniotic sacs (polyhydramnios in the recipient and oligohydramnios in the donor), and occasionally hydrops (heart failure) in one twin. The prognosis without treatment would depend on the patient's gestational age, with more than 90% of fetuses dying in previable gestations
(<25 weeks - see Lopriore E, et al.: Twin-to-twin transfusion syndrome: New perspectives. J Pediatr 1995;127:675-80). In TTTS the polyhydramnios typically causes over-expansion of the uterus, and pregnancy loss from premature rupture of the membranes or pre-term labor follows. Surviving TTTS fetuses may suffer complications related to chronic transfusion (e.g., cardiac dysfunction, brain damage), prematurity, growth restriction, and injuries related to the death of a co-twin in utero. Monochorionic twins are at higher risk (ranging from 4% to 25%) than dichorionic twins for intrauterine death of a fetus, and neurologic sequelae occur in approximately 27% of the associated surviving co-twins (see Van Heteren CF, et al.: Risk for surviving twin after death of co-twin in twin-twin transfusion syndrome. Obstet Gynecol 1998;92:215-9 ). These TTTS outcome data justify therapeutic interventions to prevent abortion, prematurity, or in utero death of one or both twins.
In addition to FLOC, treatments for previable TTTS pregnancies have included:
1. Termination of the entire pregnancy, either actively or passively (i.e., letting nature take it course). These are not recorded and rarely reported, and most patients seem to be counseled on this option.
2. Selective termination of one fetus. Modern methods that occlude the umbilical cord (PUCL) or major fetal vessels are preferred in order to avoid death or ischemic damage in the other fetus because of the twins’ chorioangiopagous status (see Challis D, et al.: Cord occlusion techniques for selective termination in monochorionic twins. J Perinat Med 1999;27:327-38, and Nicolini U, et al.: Complicated monochorionic twin pregnancies: Experience with bipolar cord coagulation. Am J Obstet Gynecol 2001;185:703-7).
3. Maternal digoxin therapy for fetal cardiac failure. This has been anecdotally successful in TTTS cases and in pump twin heart failure in acardiac twins (see Koike T, et al.: Digitalization of the mother in treating hydrops fetalis in monochorionic twin with Ebstein’s anomaly. J Perinat Med 1997;25:295-7).
4. Indomethacin therapy to curtail amniotic fluid production by decreasing fetal urine output. None of three TTTS cases (20,23 and 25 weeks' gestation) in one series seemed to respond and one of the two survivors developed multicystic encephalomalacia following a co-twin's death (see Jones J, et al.: Indomethacin in severe twin-to-twin transfusion syndrome. Am J Perinatol 1993;10:24-6).
5. Aggressive therapeutic reduction amniocentesis of the polyhydramnios is the most widely used therapy. It relieves maternal discomfort and prolongs TTTS pregnancies by reducing the risk of spontaneous rupture of the membranes. Also, by reducing intrauterine pressure, it may improve some TTTS cases by changing hemodynamics of the anastomoses. Survival rates approach 80% in some centers, but health status of surviving infants is not always reported (see Dickinson J, et al.: Obstetric and perinatal outcomes from The Australian and New Zealand Twin-Twin Transfusion Syndrome Registry. Am J Obstet Gynecol 2000;182:706-12, and Mari G, et al.: Perinatal morbidity and mortality in severe twin-twin transfusion syndrome: Results of the International Amnioreduction Registry. Am J Obstet Gynecol 2001;185:708-15. These papers report a 60 -75% survival rate with amniocentesis, but 15 -30% of survivors had significant brain abnormalities on neonatal head sonogram).
6. Amniotic septostomy (perforating the interfetal membrane septum) to equilibrate the extremes of amniotic fluid volumes seen in the sacs of TTTS fetuses (see Johnson JR, et al.: Amnioreduction versus septostomy in twin-twin transfusion syndrome. Am J Obstet Gynecol 2001;185:1044-7, and Saade G, et al.: Amniotic septostomy for the treatment of twin oligohydramnios-polyhydramnios sequence. Fetal Diagn Ther 1998;13:86-93). In these series, the septostomy to delivery interval was + 10 weeks and +81% of the twins survived. The mean age at treatment was 20 weeks (range 17-27). Despite the intentional creation of a functional monoamnionic sac for the twins by membrane disruption (and concerns that interfetal septum disruptions will lead to morbidity and mortality rates equivalent to true monoamnionic twins), no instances of amniotic band syndrome or umbilical cord entanglement were seen. Neonatal morbidity in the survivors was not reported. In recent case reports of septostomy as an iatrogenic complication of serial reduction amniocenteses for TTTS, cord entanglement was seen at delivery and one survivor had severe cystic periventricular leukomalacia of the brain.
7. Cerclage of the compromised cervix. In the last two years centers have begun sonographic measurement of cervical length in TTTS (see Skentou C, et al.: Prediction of preterm delivery in twins by cervical assessment at 23 weeks. Ultrasound Obstet Gynecol 2001;17:7-10). Cervical evaluation is now an integral part of our treatment protocol, and we consider this the ‘third variable’ of our TTTS pathophysiology/treatment paradigm. Cervical shortening or dilation, and/or funneling or internal os dilation may occur in a significant number of patients secondarily from the consequences of TTTS pathophysiology (i.e., polyhydramnios), and may lead to early pregnancy loss. In 6 of our last 24 (25%) patients having FLOC, cerclage was performed prior to transport to our center in two, prior to discharge from our center in three, and once during follow-up at home, for various degrees of distal cervical. The two patients with cervical dilation required prolonged hospitalization in addition to cerclage to achieve a successful outcome. One review of prognostic factors for laser indicated that volume of amniotic fluid drained at completion of laser (an indirect measure of uterine size) correlated significantly for delivery within the first four postoperative weeks (see Zikulnig L, et al.: Prognostic factors in severe twin-twin transfusion syndrome treated by endoscopic laser surgery. Ultrasound Obstet Gynecol 1999;14:380-7).
8. Aggressive nutritional therapy. One possible explanation for some seemingly enigmatic aspects of TTTS pathophysiology may be maternal malnutrition, i.e., hypoproteinemia and anemia (see De Lia J, et al.: Maternal metabolic abnormalities in previable twin-twin transfusion syndrome. Twin Research 2000;3:113-7). We consider maternal nutrition the ‘fourth variable’ in TTTS, and recommend nutritional supplements (e.g., 2 to 3 cans per day of Ensure or Boost High Protein) in all TTTS cases regardless of severity or invasive treatment.
9. Tocolytic therapy with various agents has been used alone and in combination with the treatments in this list.
These alternatives to surgical interruption of anastomotic vasculature are either symptomatic therapy (reduction amniocenteses, septostomy, digoxin, indomethacin, tocolytics, nutrition supplementation), or methods that create serious ethical choices and/or a nihilistic outlook for the parents and physician (pregnancy termination, feticide). Although all these treatments may increase twin survival, complications may be seen in the survivors since they fail to address the pathophysiologic processes of severe TTTS in the placenta. Comparisons among therapies are difficult as published studies have varying inclusion criteria (e.g., reduction amniocentesis studies recruit cases up to 28+ weeks, whereas we limit FLOC to gestations < 25 weeks). In addition, many patients with TTTS are managed with more that one therapy. No treatment modality has been studied in a randomized manner because of limited numbers in individual referral centers and because of the ethical issues raised by withholding treatment in a condition associated with >90% perinatal mortality rate when left untreated (see Editorial Comment in Aust NZ J Obstet Gynaecol 1995;36:16). By simply photocoagulating the anastomotic vessels directly, FLOC has the advantage of decreasing the duration of abnormal transfusion physiology in the twins, as well as providing protection from the acute interfetal hemorrhage should one twin die in utero.
Unlike most human organs approached surgically, anatomic variations represent the norm rather than exception in the placenta because no two are completely alike. In addition to the location of the placenta within the uterus, MC placentas vary in location and type of umbilical cord insertions (velamentous versus central), location of the vascular equator in the chorion plate, the type and number of anastomotic vessels and the relationship of the interfetal membrane septum to the vascular equator. The character of the anastomoses must be determined visually by fetoscopic exploration of the vascular equator. Chorioangiopagous vessels within the equator may be direct (artery to artery, vein to vein) or indirect (artery to vein) and transcotyledonary. In general placentas from classic TTTS cases are characterized by a paucity of direct A-A and V-V anastomoses when compared to MC placentas without TTTS. Conversely, in TTTS associated with significant placental asymmetry, direct anastomoses may predispose to acute transfusion and death or ischemic damage in the survivor with MC co-twin death.
The interfetal membrane septum becomes flattened on the placental surface by the varying amniotic fluid volumes of the twins with poly-/oligohydramnios sequence. It creates a relatively opaque surface between fetoscope and the chorion vasculature (see De Lia J, et al.: Placental surgery: A new frontier. Placenta 1993;14:477-85). Variations in the location and axis of the interfetal membrane septum (which is rarely seen as shown in texts centrally located in the common placenta) occur in MC placentas that are unrelated to the MC vascular equator (recall their separate embryologic origins). Therefore, the anastomotic vessels are not found directly at the base of the septum. Most often the septum is seen shifted toward and onto the donor's chorion. Occasionally, one must photocoagulate recipient ‘s vessels at the compressed septum edge to avoid leaving patent downstream communications.
The shift of the septum onto the donor's chorion has major implications for centers that used a modified FLOC technique originally described in Europe (see Ville Y, et al.: Preliminary experience with endoscopic laser surgery for severe twin-twin transfusion syndrome. N Engl J Med 1995;332:224-7, and Deprest J, et al.: Alternative technique for Nd:YAG laser coagulation in twin-to-twin transfusion syndrome with anterior placenta. Ultrasound Obstet Gynecol 1998;11:347-52). These centers photocoagulated all vessels seen crossing the interfetal membrane septum edge. Although this technique can occlude downstream anastomoses in the equator, the interruption of the vessels at the intersection of the septum and vessel also destroyed significant amounts of the donor's normal functioning placental. The technique threatened donor survival and theoretical overall survival rates for FLOC. Iatrogenic fetal losses utilizing the modified FLOC procedure may effect negatively the results of the randomized trial by the Eurofetus Group comparing the relative efficacy of FLOC versus therapeutic reduction amniocentesis.
The duration of FLOC surgery is related to the number of anastomoses (approximately eight per case), the character of the amniotic fluid (clear or murky) and location of the interfetal membrane septum. Puncture site bleeding into the amniotic fluid is the Achilles heel of fetoscopy (about 10% of surgeries) for it limits the ability to see and fire the laser. Our two channel operative fetoscope measures 3.5 mm in diameter. FLOC has proven effective in cases that were refractory to reduction amniocentesis, and with hydropic recipients that are often the target of feticidal procedures. Postoperatively the donor increases its urinary output, while the recipient's output returns to normal correcting the polyhydramnios/oligohydramnios sequence by 2-3 weeks. Recovery from the discordant fetal growth (which may be a legitimate indicator of chronic transfusion from one fetus to the other) may or may not occur after FLOC and seems to be determined by placental symmetry factors.
Outcomes of the first twenty-six FLOC procedures for TTTS performed at the University of Utah and Medical College of Wisconsin have been reported (De Lia J, et al.: Fetoscopic laser ablation of placental vessels to treat severe twin-twin transfusion syndrome. Am J Obstet Gynecol 1995;172:1202-11). That number has increased to 153 cases (most recent at St. Joseph’s Hospital - Milwaukee), with the first 33 cases considered our 'learning curve'. Starting with the 34th case, we began to perform FLOC on patients with anterior placenta implantations, which require exteriorizing the uterus.
The 34th to 100th cases were analyzed recently to determine the efficacy of FLOC subsequent to our learning curve (see De Lia JE, et al.: Previable twin-twin transfusion syndrome with fetoscopic laser surgery: Outcomes following the learning curve. J Perinat Med 1999;27:61-7). These latter 67 patients had a mean gestational age of 21.1 +1.7 weeks (range 18 - 24.5) with a mean fundal height of 33.1 cm when treated. Eighteen (27%) had failed another treatment method before FLOC. All 67 cases delivered with 82% (55/67) having at least one surviving twin and 92/134 (69%) of the twins surviving overall. Thirty-seven had surviving twins, 18 had one survivor (6 neonatal and 12 fetal deaths), and 12 had none. The mean duration of pregnancy following FLOC was 9.9 +5.5 weeks. Only 4 of 93 (4.3%) survivors had significant handicaps at a mean follow-up of 60 months (range 48 - 84). In the last two years 90% have at least one survivor, overall survivor rate is 75%, pregnancy is prolonged 11.1 weeks, and neurologic handicap rate is < 2% (improvements we credit to cervical evaluation and cerclage when indicated).
The data from colleagues in Europe indicate somewhat similar infant outcomes and survival rates for FLOC - 73% at least one survivor and 55% overall and neurologic injury in 4.2% (see Ville Y, et al.: Endoscopic laser coagulation in the management of severe twin-to-twin transfusion syndrome. Brit J Obstet Gynaecol 1998;105:446-53). However, the photocoagulation of vessels at the membrane septum edge rather than anastomoses in the equator may explain why fewer cases have both fetuses survive in Europe (36%) when compared to our cases (65%). In addition, their rate of neurologic handicap is higher (see Sutcliffe AG, et al.: Outcome for children born after in utero laser ablation therapy for severe twin-to-twin transfusion syndrome. . Brit J Obstet Gynaecol 2001;108:1246-50). This may reflect incomplete separation of the twins’ circulations from by the limitations posed by a percutaneous approach, rather than laparotomy as preferred in our center.
Technical challenges of FLOC include bleeding from the fetoscope insertion site, stained amniotic fluid (in patients both with and without a history of previous invasive procedures), placental surface distortions or placentas that involved two uterine surfaces. Most but not all of the surviving twins with significant morbidity came from these cases. Two couples elected termination of pregnancy within 2 weeks after an incomplete FLOC, both of which were complicated by the immediate postoperative death of one fetus and possible cerebral changes in the surviving fetus on sonogram. We had one case in which both fetuses survived with congenital anophthalmia not thought to be a TTTS related anomaly. Significant maternal complications included one spontaneous, complete lung atelectasis (anesthetic complication) in a patient with a history of asthma that resolved, and one patient (No. 102) who had a wound disruption (incision separation) severe enough to warrant pregnancy termination to effect repair and healing. Maternal deaths have occurred after FLOC in one U.S. center and two European centers, with the two European deaths remote from surgery (one case each of severe toxemia and placenta increta). Metabolically, all women admitted for FLOC at our center demonstrated hypoproteinemia and anemia (see above). The low colloid osmotic pressure from reduced serum albumin may explain pulmonary edema reported after fetal therapy or treatment of pre-term labor in multiple gestation.
Twin-twin transfusion syndrome remains a severe complication of monochorionic twinning and one of the most challenging in contemporary perinatology. We use a dynamic TTTS paradigm with four components, two fetal-placental (the anastomoses and placenta sharing) and two maternal (nutrition and cervical status). Although there are several treatment options available for previable TTTS which result in a majority of twins surviving and intact, FLOC is the only treatment method which can limit the duration of severe TTTS fetal pathophysiology (see Pietrantoni M, et al.: Mortality conference: Twin-to-twin transfusion. J Pediatr 1998;132:1071-6). Although survival rates and treatment to delivery intervals may be comparable, the photocoagulation of all vascular anastomoses in the vascular equator results in less neonatal morbidity than other treatment methods. Comparative trials for different treatment modalities are underway (see Hecker K, et al.: Endoscopic surgery versus serial amniocenteses in the treatment of severe twin-twin transfusion syndrome. Am J Obstet Gynecol 1999;180:717-24, and Johnson JR, et al.: Amnioreduction versus septostomy in twin-twin transfusion syndrome. Am J Obstet Gynecol 2001;185:1044-7). Given the nature of the placental abnormalities and developmental anomalies in MC twins, survival of all twins and elimination of all risk for neurologic handicap in survivors may be impossible. Ultimately, clinicians and parents alone must decide what perinatal mortality rates and long-term morbidities are acceptable when choosing a TTTS therapy.
Baldwin VJ. Pathology of Multiple Pregnancy. Springer-Verlag, New York, 1994. P.199.
Benirschke K, Kaufmann P: Pathology of the Human Placenta. 4th ed. Springer-Verlag, New York, 2000, ISBN 0-387-98894-7