Summary of the history and pathophysiology
The pathophysiology of twin-to-twin transfusion syndrome (TTTS) is complex, and its understanding has evolved along with the evolution and revolution of fetal diagnostic and therapeutic techniques. The shunting of blood from donor to recipient is but an oversimplified way of describing the intricate hemodynamic changes that take place in a pair of monochorionic fetuses.
The most essential components of the currently accepted hypotheses are the vascular anastomoses on the chorionic plate (Figs. 1–2). Histologically, three types have been described: artery-to-artery (AA), artery-to-vein, and vein-to-vein anastomoses. AA and vein-to-vein are bidirectional, superficial, and indirectly connect the two fetal umbilical cords.1 However, the artery-to-vein connections are deep, with unidirectional flow.2 Further histopathological examination reveals that these connections are in fact cotyledons with afferents from a chorionic artery of one fetus and efferents to a chorionic vein of the other. Blood flow is unidirectional and at a gentler pace through these cotyledons.3
Hence, it is essential to intervene and correct the imbalance before the progression to a vicious cycle of loss of homeostasis leading to demise of both affected fetuses (Fig. 3).
The spontaneity in the onset and resolution of acute TTTS can be explained by the dynamically changing functional placental anatomy which can occur because of thrombosis of a vascular communication or infarction of a placental region that may have contained an anastomosis.
A multicenter European randomized clinical trial has been historical in demonstrating the superiority of fetoscopic laser photocoagulation (FLP) over amnio reduction for the treatment of the syndrome. Photocoagulation was associated with improved neonatal survival and reduced neurologic injury. The results of this pioneering trial have revolutionized the patient care and fetal survival rates in pregnancies complicated by TTTS.7
First-trimester diagnosis of TTTS – current protocols
The chorionicity of the pregnancy will determine if the pregnancy has the potential to be complicated by TTTS. Twinning after differentiation of the chorion but before the amnion results in a monochorionic diamniotic twin gestation which is the placental type commonly seen in TTTS.8
As a rule, the monochorionic pregnancies are monozygotic. Rarely, fusion of separately fertilized embryos can result in the trophoblasts intermingling to form a single placenta – only a handful of such dizygotic monochorionic twin pregnancies have been described in the literature till now.9,10
The key to monitoring a monochorionic pregnancy for potential TTTS and therapeutic interventions is the early recognition of the monochorionicity. The ultrasound diagnosis of chorionicity is technically very easy during early gestation but becomes more difficult as the pregnancy advances. Hence, vigilant identification and documentation of the chorionicity of a twin pregnancy is important to facilitate the early diagnosis and therapy of TTTS which is a risk for the monochorionic group.11
Gold standard: Imaging of the intertwin membrane insertion on the chorionic plate in the first-trimester. The insertion will appear as the shape of “T” in the monochorionic or as “λ” in the dichorionic.12
In case the first-trimester ultrasound is not available or was not performed, a high-resolution probe can be used for the imaging of the membrane, and the number of layers of the membrane can be suggestive of the chorionicity. However, this is method has low sensitivity and high interobserver variation, hence it should be considered as a method of last resort.12
The rule of thumb is: diagnosis of chorionicity gets more difficult and less reliable as the pregnancy progresses over 14 weeks. The simplest diagnosis would be in the early first-trimester, where two distinct amniotic sacs within the same chorion would be visualized easily by transvaginal ultrasound.11
Surveillance protocol for monochorionic pregnancies at risk of TTTS
The diagnosis of TTTS is not clinical or biological, but it is based on the ultrasound findings. Ultrasound surveillance plays an important role for timely detection and treatment of TTTS.11,12
It is very essential that monochorionicity has been indisputably established during the first-trimester ultrasound, for a pregnancy to be considered at risk for development of TTTS. Once this has been confirmed, established guidelines recommend surveillance imaging to start at 16 weeks, with scans repeated every 2 weeks thereafter for the monitoring of parameters suggestive of TTTS (Fig. 4). This includes, for both fetuses11,12:
- (1) Amniotic fluid volume (preferred technique – deepest vertical pocket: DVP).
- (2) Bladder status – filled (visible) or empty (not seen).
- (3) Doppler indices – paravesical umbilical artery pulsatility index, middle cerebral artery pulsatility index (MCA-PI), and middle cerebral artery peak systolic velocity (MCA-PSV), ductus venosus pulsatility index.
- (4) Membrane infolding as an indirect sign of fluid imbalance.
- (5) Growth parameters including estimated fetal weight (EFW).
In situations where amniotic fluid discordance is noted but does not fulfill the diagnostic criteria of TTTS, the surveillance frequency can be increased to weekly scans to ensure that the discordance is not progressive (Fig. 4).
These are additional points of surveillance, and they are performed in addition to the routine ultrasound protocol for fetal abnormalities.
Diagnosis of TTTS – staging and prognostication
The single, essential criteria for the diagnosis of TTTS is significant amniotic fluid volume discordance between the two fetal sacs. Conventionally, the ultrasound parameter used is the DVP of amniotic fluid in each sac. The DVP is defined as the largest vertical pocket of amniotic fluid without the presence of umbilical cord or other fetal parts.13
The diagnostic cut-off is: DVP value of <2 cm in “Donor” sac – labeled as oligohydramnios and DVP value of >8 cm before 20 weeks or >10 cm thereafter in “Recipient” sac – labeled as Polyhydramnios.13
Once the diagnosis of TTTS has been established using this criterion, it is essential to refer the woman to a specialist tertiary care center which can: (1) deal with further assessment, staging, and prognostication of TTTS, (2) provide therapeutic corrections if indicated.
It is necessary that this referral and the ultrasound assessment are done as expeditiously as possible, to be able to plan timely intervention and ensure the maximal chances of survival for both the fetuses.
The Quintero staging system has been used for the last 20 years for the classification of TTTS.14 It is worth mentioning that this staging system does not accurately predict the outcome and is not a representation of the chronological evolution of the disease. For example, Stage I is not necessarily associated with the best outcomes, and it can become Stage 5 without passing through Stages 2, 3, and 4. Despite these limitations, the Quintero staging system remains the classification system of choice as the stages may not represent different times during the natural history of the disease, but rather different forms of presentation with possibly different prognosis.12
- Stage 1: Amniotic fluid discordance fulfilling the diagnostic definition.
- Stage 2: Bladder not visualized in the donor twin.
- Stage 3: Critically abnormal Doppler indices in one or both twins (Stage 3-Donor or Stage 3-Recipient).
- (1) Absent or reversed umbilical artery diastolic flow.
- (2) Reversed ductus venosus a-wave flow.
- (3) Pulsatile umbilical vein flow.
- Stage 4: Hydrops in one or both twins.
- Stage 5: Demise of one or both twins.
Additional staging systems from Children's Hospital of Philadelphia, USA, (Children's Hospital of Philadelphia cardiovascular score)15 and Cincinnati Children's Hospital, USA (Cincinnati cardiovascular score)16 have been proposed and debated, but the Quintero system is the most commonly used due to the ease of imaging. These classifications incorporate mainly the additional cardiovascular parameters of the disease, but studies have not conclusively proven their improved prediction of outcome following treatment (Fig. 5).17,18
Non-laser modalities of treatment – benefits and pitfalls
Over the last few decades, several nonlaser therapeutic interventions have been researched for improving the outcomes in TTTS. These include selective feticide, amnio reduction, septostomy, early delivery in the third-trimester, medical therapies such as digoxin and prostaglandin synthase inhibitors, intrauterine transfusion, and so on. What follows is a brief discussion of the popular nonlaser modalities and their place in the current practice.
Amniodrainage is the procedure of reduction in the volume of amniotic fluid in the polyhydramnios sac by percutaneous needle aspiration. It was the first widely employed intervention and it still has a place as an adjunct therapy, performed simultaneously with laser photocoagulation.
The rationale behind amniodrainage is the fact that increased amniotic fluid pressure adversely impact fetal partial pressure of oxygen independent of gestational age.19 There is a significant association of elevated amniotic fluid pressure with abnormal fetal blood gas status and impaired uteroplacental perfusion.20,21 Also, polyhydramnios may cause secondary maternal respiratory compromise, preterm labor, or preterm membrane rupture.22
Amniodrainage does not alter the underlying pathophysiology of the disease but is rather a treatment for the resultant symptoms of the pathology. As a result, it plays a small, albeit important role as an adjunctive treatment to laser therapy for immediate symptomatic improvement in the uteroplacental circulation, delaying development of preterm labor, and relief of maternal discomfort due to polyhydramnios.
However, the superiority of laser therapy over amniodrainage has been conclusively established by scientific evidence,7,23 and there are very few indications of isolated serial amniodrainage in the current therapeutic regimen for TTTS. The most important indications are: treatment when laser surgery is not available or not easily accessible12 and symptomatic relief in Stage I disease in third-trimester cases (vs. preterm delivery).11
As a part of the conventional protocol, the average volume of amniodrainage in a single sitting is recommended to be not more than 3 000 mL over 30 minutes. This has been proven to provide the maximal symptomatic benefit with lower probability of the gravest complication of placental detachment due to sudden fall in amniotic fluid pressure.24
Selective fetal reduction
Selective fetal reduction in TTTS involves occlusion of the umbilical cord of one of the fetuses with the hope that the co-twin will survive intact. By ablating the vascular communications between the fetuses at the level of the umbilical cord, it is possible to prevent feto-fetal hemorrhage and prevent further hemodynamic damage to the surviving twin. Selective fetal reduction can be of two types: primary or secondary.
Primary reduction is done without an attempted laser treatment for TTTS, most commonly for two indications: parental choice or discordant anomalous twin. Rarely, it is necessary to do a secondary fetal reduction as a modality of last resort in severe TTTS cases where laser therapy has failed to improve the situation, severe lesions are objectivized in one twin, or technical difficulties limit the performance of a complete laser photocoagulation.25 It is advisable to discuss consent for fetal reduction along with the consent for laser photocoagulation in cases where conservative placental surgery would prove impossible intraoperatively, so that the procedure can be performed at once and the patient would not require a second fetoscopy.
In our practice, we perform selective fetal reduction by complete laser coagulation of the umbilical cord of either fetus under direct fetoscopic visualization, using diode laser energy at 40 watts before 18–20 weeks. Beyond 20 weeks we prefer to use bipolar forceps under direct ultrasound guidance (Fig. 6). The procedure is technically simple, though a bit tedious on appearance. The benefits of using fetoscopy are ensuring the complete coagulation of the cord but also to associate laser photocoagulation of the rooting vessels of the umbilical cord before occlusion with bipolar forceps using the same trocar. In monoamniotic pregnancies, cord coagulation is followed by cord transection using a laser fiber under direct fetoscopic control. This ensures de-tethering of the reduced fetus so that cord entanglement is avoided, and the surviving fetus can develop without any spatial restriction for the remainder of the pregnancy.
With increasing evidence of the safety and efficacy of laser therapy,26,27 conservative management has a very limited role to play in the current therapeutic scenario for TTTS. Coupled with the knowledge that the staging is nonchronological, Stage I pregnancies are also likely to suffer from poor outcomes with respect to progression of disease and reduction in take-home baby rates. In such a setting, conservative approach is indicated only in cases with very early, uncomplicated Stage I disease without any associated poor prognostic signs, for example, cardiac dysfunction.
In the absence of maternal symptoms and a long cervix, 50% of Stage I cases will remain stable or regress. However, the other half will progress within 1 or 2 weeks towards higher stage TTTS or maternal symptoms indicating surgery. There is no advantage to initial surveillance versus immediate surgery. Provisionally, weekly surveillance can be offered (unpublished data).
For cases with late-developing IIIrd trimester TTTS (beyond 26 weeks) laser surgery may carry a risk equivalent to the risk of complications faced by a preterm neonate. For such cases, the strategy depends upon surgical access of the placental surface. The procedure can be planned as described below and fetoscopic visualization will assess the chances of success of surgery. When those are anticipated to be good, photocoagulation is performed. However, when surgery is likely to be incomplete, then amnioreduction is performed alone through the operative trocar and elective preterm delivery becomes the treatment of choice. Up to the planned preterm delivery, these cases are managed conservatively, to administer corticosteroids for lung maturation and magnesium sulfate forneuroprotection.11
Septostomy is a therapeutic technique that is obsolete in the current practice setting and has been abandoned by all TTTS treatment centers across the world. The concept behind a septostomy was to open communication in the inter-twin membrane between the two amniotic sacs. Theoretically, it would allow to relatively balance the quantities of fluid between the two amniotic cavities. Practically; however, isolated septostomy has been a spectacular failure at alleviating the symptoms or the progression of TTTS. When the septostomy is large, it also has the disadvantage to transform a diamniotic pregnancy into a pseudo-monoamniotic, with all the attendant complications of a monoamniotic pregnancy.11,28
FLP – biophysics and mechanics of laser therapy, basics of laser safety
When described simplistically, the idea behind laser photocoagulation for TTTS is the interruption of all vascular anastomoses on the common chorionic plate. This results in the separation of the circulations shared by both fetuses and termination of the imbalanced exchange of blood between them. Ideally, all placental vascular anastomoses are to be identified and obliterated for a successful dichorionization. However, this may not be possible in all cases.
The main component of this procedure is laser energy. Energy generated by different laser sources has different characteristics, including depth of penetration, tissue absorption, traversal through various media, and energy dissemination. To be able to use laser energy for TTTS, the following criteria must be fulfilled:
- (1) Peak absorption spectrum appropriate for target tissue: blood vessels, that is, hemoglobin.
- (2) Selectivity for the target tissue, to minimize damage to surrounding structures.
- (3) Minimal diffusion, maximal thermal conversion at a point target of the tissue.
- (4) Low absorption for water to prevent heating up of the amniotic fluid – important as the laser beam must travel through the amniotic fluid to reach the target.
- (5) Ability for the laser beam to be propagated via optic fiber which can be utilized endoscopically.
Currently, two lasers exist which are compatible with the above criteria: neodymium – yttrium aluminum garnet (Nd:YAG) laser (1 064 nm) and diode laser (940 nm). Even though theoretically the diode laser is closer to the hemoglobin absorption spectrum, practically the therapeutic efficacy of both the lasers are equivalent. The diode laser is used more commonly, because of its smaller size and lower cost.29
These lasers, being class four devices, should be utilized only after acquired proper training and familiarity with the equipment.30 It is necessary to ensure that all the laser safety rules are followed correctly, before and during the laser procedure. The aim is to protect all individuals (including patient and staff) from exposure to visible or invisible laser radiation – direct, refracted or reflected. The most important component of laser protection during TTTS surgery is utilization of filtered eyeglasses, specific for the wavelength of the laser beam in use.
FLP – pre-procedure workup
After the confirmation of the diagnosis and stage of TTTS, a comprehensive case review is performed to eliminate any overlooked ultrasound abnormalities and formalize the obstetric case history. The patient is assessed by the preoperative unit, which includes blood testing for complete blood count, type and screen, coagulation profile, electrolytes, liver enzymes, blood urea nitrogen, and creatinine.31 A consult is arranged on a priority basis with an anesthesiologist for evaluation.32
To the patient and her family, TTTS presents the difficulty of dealing with the disease as well as with the psychological and emotional issues attached to it, including potential fetal demise and pregnancy loss.33 Hence, counseling and informed consent are the most essential components of the pre-procedure stage of care. The designated nurse-midwife coordinates the initial consult of the patient, and is the principle point of contact during the entire period of the assessment and treatment. Whenever possible, it is recommended to have a multi-disciplinary counseling session which should involve the treating fetal specialist, obstetrician, anesthesiologist, and neonatologist. It is essential, with the patient's permission, to include her immediate family in the counseling session. The fetal specialist explains regarding the treatment options, the indications for FLP and describes in brief regarding the procedure and what to expect during the day of the surgery. The multi-disciplinary team then participates in the discussion. Any queries from the patient or her family regarding the procedure or its complications are appropriately explained and informed consent for the procedure is acquired.31
FLP – preoperative checklist
FLP is usually an emergent procedure, hence the lead time to performing the surgery is quite short. The pre-operative checklist is typically divided into three parts (Fig. 7):
With rapid advancement in anesthetic techniques and availability of safer drugs, spinal or epidural anesthesia is now an acceptable option for FLP in patients with anxiety or lower back pain tolerance. Epidural anesthesia is preferred only in cases which require a cervical cerclage to be placed simultaneously. However, local anesthesia remains the modality of choice for these procedures.32 The drug of choice is 1% plain lignocaine (without preservatives) and up to 30 cc can be infiltrated under ultrasound guidance, extending till the myometrium along the pre-decided path of the fetoscope sheath. Additional intravenous medication for conscious sedation, for example, midazolam can be considered at the discretion of the anesthesiologist (Fig. 8).
Preoperative ultrasound assessment plays a crucial role in determining the site of entry for the fetoscope. Before commencement of the procedure, the placental anatomy is mapped on the maternal abdomen, highlighting the approximate locations of both the cord insertions. In majority of cases, the location and direction of the inter-twin membrane can be gauged by the lie of the donor twin. It is important to note the location of the intertwin membrane, to avoid the consequences of accidental damage to the membrane. The two basic principles to decide a point of entry are29: (1) the scope, when inserted, should be at a direction perpendicular to the direction of the intertwin membrane, and consequently perpendicular to the lie of the donor twin. This will enable easy access to the entire membrane length over the chorionic plate and minimize the amplitude of scope movements necessary to complete the procedure; (2) the direction of scope at entry should be aligned with a virtual line drawn between the two cord insertions previously assessed. This would again provide easier access to all the anastomoses (Fig. 9).
For all insertions, care is taken to avoid injuring the superficial epigastric vessels as well as the deep uterine vessels, by using either the midline or lateral region of the abdomen. Unless absolutely necessary, it is preferred to insert the scope in an area devoid of placenta in cases where the placenta is anteriorly placed.
FLP – steps of the procedure
We describe here a summary of the surgical protocol for FLP, after choosing the point of entry and infiltration of local anesthetic agent.
Entering the amniotic cavity
A small, 1 cm stab incision over the maternal skin is performed using a scalpel. The depth of the incision must be carefully controlled, to avoid injuring the maternal myometrium, especially in patients with a thin abdominal wall.
Two options exist for entry into the amniotic cavity of the recipient twin – reusable metal trocar and cannula or Seldinger technique with a disposable sheath. We prefer the latter because of the simplicity and lower potential for tissue damage during sheath insertion, as the entry is under direct ultrasound visualization.
The Seldinger technique involves entering the amniotic cavity with an 18G echotip needle under direct ultrasound vision. Once inside, the spring wire of the sheath is threaded through the needle in such a way that about one-third of the length of the wire is extra-abdominal, and the needle is removed. Now, the sheath along with its rigid introducer is threaded onto the spring wire and pushed through the abdominal wall and myometrium with gentle but firm pressure till it is visualized to be entering the amniotic cavity. The introducer and the wire are then removed. It must be reiterated that the entire procedure is carried out under continuous ultrasound vision.
Once the sheath is in place, amniotic fluid sample is collected if the patient has consented for genetic testing. The first 5–10 mL of fluid through the irrigation channel of the sheath is aspirated and discarded. Following that, around 30 cc to 50 cc (depending on the requirement for testing) of fluid is aspirated using a 50 cc syringe.
The fetoscope is the most important piece of equipment for this surgery. We use the customized semi-rigid fetoscopes for laser surgery which are 30 cm in length and in different sizes – the routine 2 mm and smaller 1.2 mm for use at earlier gestational ages (less than 20 weeks). These fetoscopes can be either straight (Karl Storz 11630 AA +11630 KF 0o 2 mm; Karl Storz, Tuttlingen, Germany) or curved (for anterior placentas, Karl Storz 11630 AA+ 11630KF curved 0o 2 mm; Karl Storz) (Fig. 8).34
The appropriate fetoscope is chosen for the case, and inserted through the sheath, into the amniotic cavity. The lens system of the fetoscope is connected to a current-generation high-definition camera system which has recording capabilities. A cold light source (xenon in our setup) is used to provide neutral-tone lighting. Sterile operative field is maintained by use of sterile polythene covers with connectors for the camera and light source cables.
The 600 μm or 400 μm laser fiber is passed through the operating channel of the fetoscope, with a side-firing configuration. The fiber is connected to a diode laser, configured at 40 watts of power. Till the initial survey is completed, the laser is unarmed.
If amnioinfusion is required, the irrigation channel of the sheath is connected to an infusion set and warmed sterile saline is pushed using a 50 cc syringe and a three-way connector.
Identification and coagulation of anastomoses
Once the fetoscopic view has been established, the next step is the inspection of the vascular architecture of the monochorionic plate. The location of both the placental cord insertions are visualized, and the direction of the inter-twin membrane is identified. Due to fewer layers on a microscopic level, the membrane of monochorionic pregnancies is quite thin and translucent, thus permitting the visualization of the cord insertion and vascular anatomy of the donor sac. The polyhydramnios of the recipient sac tends to push the membrane towards the donor fetus (creating the “stuck twin” moniker), hence the vascular equator is not usually located at the same level as the membrane. It is important to identify the vascular equator as a relatively avascular region of the placenta except for the anastomosing vessels between the two fetuses. It is about a few centimeters in width and can be parallel or perpendicular to the axis of membrane insertion. The deep vascular anastomoses in this region can be easily identified by noting that the perfusing artery and vein belong to different fetuses. Having precisely located the two cord insertions allows for a quick and reliable identification of the cord to which the examined anastomotic vessels belong to. This largely prevents the need to unnecessarily trace the vessels back to the cord insertion.
The complete region of the vascular equator is explored, from one placental edge to the other, and the culprit anastomoses identified. In our surgical practice, we follow the selective photocoagulation protocol where we recognize and simultaneously coagulate each anastomosis as we go along the vascular equator from one end to the other. Coagulation is by the no-touch technique. The laser is armed, and the laser shot is activated by the surgeon using the foot-switch, with the tip of the fiber at an approximate 1 cm distance above the vessel, with the laser beam perpendicular to the chorionic plate for the maximal transmission of energy. Generally, about 3 or 4 laser shots are required for the complete coagulation of an average-sized vessel, evidenced by the visible blanching or whitening of the target region. For larger vessels, we begin the coagulation from the edges of the vessel and proceed towards the center. Large vessels require more laser shots for complete coagulation.
After all the identifiable anastomoses have been photocoagulated, we carry out complete dichorionization of the placenta using the Solomon technique.29,35 This helps include the smaller vessels which were missed and decreases the potential for post-laser complications or the need for a repeat procedure. The technique is very simple – using continuous laser power, the whitened “dots” of the already coagulated vessels are connected by coagulation of a thin strip of chorionic plate between them. Thus, all the “dots” are joined, resulting in the complete coagulation at the vascular equator and the dichorionization of the placental circulation.
A review is done of the coagulated vessels, to double-check the completeness of coagulation, and confirm that no vessels have been missed. Occasionally, spastic vessels may initially give a mistaken appearance of a successful coagulation. These vessels can be identified on a review, and the coagulation can be completed.
After the completion of coagulation, the fetoscope is removed, and amniodrainage is done. This serves to decrease the hydrostatic pressure on the placenta as well as cervix, thus improving the placental blood flow, and reducing the chances of preterm labor. Anecdotally, placental abruption has been mentioned as a complication of large volume amnioreduction. Hence, gradual amniodrainage is done, maximum up to 3 L of fluid in one sitting. Amniodrainage is done via a sterile tube connected to the fluid channel of the sheath and an electric suction machine connected to a volume-graded flask for measurement. The fluid level is monitored on ultrasound, and drainage is stopped once the SVP reaches 5 cm–6 cm. of the donor sac.30
Once amniodrainage is complete, the rigid introducer is inserted into the sheath, and the sheath is removed in a swift motion under ultrasound vision, making sure to avoid further trauma to the membranes or the surrounding viscera. Hemostasis is confirmed, and the skin incision is closed with a single stitch using absorbable sutures.
Improvisation in case of fluid turbidity
Fluid turbidity can be encountered due to several causes, e.g. advanced gestational age, excessive vernix, bloody discoloration due to previous procedures like amniocentesis, cordocentesis, and so on. In these situations, the fetoscopic vision is grossly impeded, making the performance of the coagulation difficult. In such a situation, amniotic fluid exchange is an excellent option. Using the fluid channel of the sheath and a three-way connector, about 500 mL at a time of turbid amniotic fluid is aspirated using the suction tube, and an equal amount of warmed saline is infused. This can be repeated multiple times, till the quality of the aspirated fluid is satisfactorily clear.30
Improvisation in case of vascular equator extending to donor twin sac
In many cases, the vascular equator is found to extend beyond the inter-twin membrane, into the donor amniotic sac. This can be treated in two ways, depending on the extent of the anastomoses in the donor sac.
(1) If the anastomoses are close to the membrane insertion, and can be visualized from the recipient sac, laser photocoagulation can be carried out from the recipient sac by transmitting the laser beam through the membrane, targeting the anastomosis. By principle, as the membrane is avascular, it will not absorb the laser energy, which will be transmitted to the donor sac and the culprit anastomoses can be coagulated while keeping the membrane intact.
(2) If the anastomoses are beyond the reach of such laser treatment, the scope is pushed through the inter-twin membrane to explore the donor side and subsequently complete the coagulation process. This is achieved in two steps: first the membrane is pierced by a sharp push of the laser fiber under fetoscopic control, while keeping the fetoscope steady. Once the laser fiber is inside the donor sac, it functions as a guide-wire and the rest of the scope is introduced across the membrane over this guide-wire.
FLP – protocol for post-laser follow-up
The end-procedure ultrasound includes documentation of the fetal cardiac activity and single largest pocket of amniotic fluid for both the fetuses.
After a minimum period of 24-hour postsurgery observation for maternal complications, the patient can be allowed to continue the remainder of her follow-up as an outpatient. The postlaser fetal follow-up ultrasound assessment is carried out every twelve hours for a total of four examinations postoperatively.29
The parameters to be assessed are: (1) fetal cardiac activity; (2) reassessment of TTTS staging for progression or regression – SVP of amniotic fluid, donor bladder visualization, Doppler flow assessment; (3) measurement of MCA-PSV and calculation of the multiple-of-median (MoM) values – for evolving twin anemia polycythemia sequence (TAPS).
If the post-laser evaluations are satisfactory, further follow-up scans are scheduled weekly, and fetal growth is also assessed, along with weekly assessment of the above parameters.12
Post-FLP complications (Table 1)
Preterm premature rupture of membranes
A common post-procedure complication, characterized by a sharply delineated defect at the site of trocar insertion, of a size corresponding to the size of the instrument used.40
Even though a multitude of studies are currently underway to create a membrane sealing device, none of them are at a stage of clinical utility.41 The current approach to preterm premature rupture of membranes is conservative follow-up with antibiotic prophylaxis, steroid for fetal lung maturation and delivery in situations of suspected chorioamnionitis or onset of preterm labor.
Post-surgery, the mean gestational age at delivery is 32–33 weeks. Usually, preterm labor is associated with already ruptured membranes. However, the occurrence of spontaneous preterm labor is 1% and, in these cases, tocolytics can be used to delay the delivery for up to 48 hours to ensure fetal lung maturation after steroid administration.
Maternal complications (bleeding, ascites)
From the maternal point of view, FLP is a very safe procedure. Rarely, complications like maternal bleeding or iatrogenic amniotic fluid ascites may occur. Bleeding can be controlled by pressure hemostasis. In case of severe uncontrolled bleeding, a laparotomy for myometrial suturing may be indicated. Iatrogenic ascites is usually self-resolving, and just necessitates follow-up monitoring of the patient.
Residual anastomoses: recurrence of TTTS, TAPS
Even after FLP, there is a possibility of missed anastomoses which can cause complications as the pregnancy evolves. By performing dichorionization using the Solomon technique, the incidence of these residual anastomoses can be reduced.12,35,42,43
Anastomoses are likely to be missed during fetoscopic laser most commonly due to the following reasons:43,44 (1) microscopic size, not visible on fetoscopic view; (2) compressed vessels due to the hydrostatic pressure of polyhydramnios, which regain their vascular flow after amnioreduction; (3) connections located beyond the field of view of the fetoscope, hence not visualized; (4) inadequate coagulation of anastomoses, resulting in revascularization and resumption of blood flow.
Around 10%–20% of FLP cases are likely to suffer from the consequences of residual anastomoses.11,12,45 These include:
Recurrence of TTTS
TTTS can recur in about 10% of cases, and persistent large-caliber vascular connections are suspected to responsible for this.4,46 Hence, it is mandatory to adhere to the weekly surveillance protocol even after FLP till the end of the pregnancy, to ensure timely diagnosis and management. Treatment options for recurrence include repeat laser surgery, amniodrainage, selective fetal reduction or preterm delivery.45 A repeat laser surgery is more complex. The technical limitations which might have resulted in incomplete coagulation during the first FLP would still hinder the second FLP, and it would be difficult to achieve a complete resolution of TTTS.11
When miniscule vascular connections exist between the monochorionic pair of twins, imbalanced transfer of slow-flowing erythrocyte-rich blood leads to a characteristic syndrome known as TAPS.4,12,46 TAPS can occur spontaneously, alongside TTTS. Or it can develop after FLP, when large connections are coagulated but the smaller vessels are missed. In the former, the Donor is anemic, although in the latter situation it is the recipient that is virtually always the anemic twin.
Noninvasive diagnosis of fetal anemia is possible by Doppler assessment of the MCA-PSV and calculation of MoM values based on the gestational age of the fetus. The measurement of the MCA PSV should be technically accurate, measured during fetal quiescence, interrogating the correct location on the MCA with minimum angle correction. TAPS is diagnosed when the anemic fetus has an MCA PSV MoM value above 1.5 and the polycythemic fetus has MCA PSV MoM below 1.0. The staging of TAPS is as described by Slaghekke47 and Lopirore.48
The incidence of spontaneous TAPS is 5%.12 However, post-laser TAPS incidence can be as high as 13%.45 Fetuses affected with TAPS are at risk of serious consequences, which include neurological damage and fetal demise for the anemic donor fetus, and cardiac failure and hydrops for the polycythemic recipient fetus. Treatment options for TAPS include repeat laser for complete dichorionization, or intrauterine fetal transfusion for the anemic donor fetus. The management of TAPS is individualized to each pregnancy, depending on the severity, gestational age, technical limitations and choice of the couple.12,29
Complications specific to FLP
The imbalanced circulation of monochorionic fetuses predisposes them to the risk of ischemic or hemorrhagic neurological damage.49 This can be spontaneous, as a part of the evolution of the TTTS pathology. FLP has the potential to reduce the incidence for such complications.49 However, postlaser neurological accidents are possible, especially in cases which are technically difficult, or with residual vascular connections. Single fetal demise can also lead to sudden central nervous system damage, with the dead fetus acting as a low-pressure receptacle of blood, leading to ischemia in the survivor.50,51
Varying reports in literature peg the incidence of neurological complications for monochorionic pregnancies as 5%–10%.49 The most common signs on imaging include ventriculomegaly, periventricular hyperechogenicity or cysts, altered cerebral development, and disturbed neuronal migration.11,49 Ultrasound surveillance is complemented by magnetic resonance imaging (MRI), which can aid in more detailed examination of the fetal brain. An MRI at 30–32 weeks is recommended for all complicated monochorionic pregnancies.52
Pseudo-amniotic band formation
Complex FLP which includes interaction with the inter-twin membrane, or traversal of the membrane for entering the donor sac, has the potential to cause a rare complication of pseudo-amniotic bands. The incidence is 1%–2%.11,53,54 The damaged membrane acts as a constricting force on a part of a fetal limb, most commonly fingers or toes. The highest risk situation follows the post-operative demise of one twin. If identified at an early stage, laser ablation of the constricting membrane can restore the circulation and avert any long-term damage.29 But, persistent restriction of vascular supply may lead to autoamputation in-utero, requiring postnatal orthopedic management for the affected limb.
Single fetal demise in a monochorionic pair
Fetal demise can be predicted to occur as the pathology of TTTS progresses. However, the sequence of progression of TTTS is not necessarily stage-wise. As mentioned previously, the functional vasculature of a monochorionic placenta is dynamically changing. Thromboses or infarcts of the placental vessels may cause sudden hemodynamic changes in either one or both fetuses. The resulting severe ischemia may lead to a sudden demise of the fetus without any Doppler changes suggestive of imminent demise.
When the diagnosis of single fetal demise is made within a few days of the accident, assessment of the survivor is a thorough search for signs of anemia (hydrops, hyperechogenic bowel, ascites, dilated heart, but mainly through measurements of MCA-PSV). Early correction of severe anemia by intrauterine transfusion is life-saving, since demise of the second twin follows in up to 40% of single demise cases. However, the anticipated benefit in preventing secondary hypoxic-ischemic brain lesions is more hypothetical. Emergent delivery of the surviving twin is not a sensible option since it will precipitate the emergent management of an unplanned (extremely) preterm and anemic neonate. In addition, delivery will not prevent brain lesions to develop and once the baby is born there will be no alternative management option.
Away from the acute event of the fetal demise, surveillance by ultrasound may uncover hypoxic-ischemic brain lesions in 20%–40% of cases. MRI is a useful adjunct to ultrasound from 28–30 weeks, especially to diagnose neuronal migration abnormalities. Brain lesions following intra-uterine demise of the co-twin may take up to 8 weeks to become amenable to prenatal imaging.55
Peripartum and postpartum care for post-FLP pregnancies
The post-FLP follow-up of the pregnancy continues with weekly assessment as per the surveillance protocol. For non-complicated pregnancies, the gestation can be allowed to progress to 34–35 weeks, although some groups would allow this to proceed to 37 weeks. Although delivery by cesarean section is mainly for obstetric indications around 80% of our cases are born by cesarean section either following parental choice or for an immature cervix, malpresentation or significant growth restriction of one or both twins. The choice of an elective moderate prematurity is based upon the comparison of the morbidity of late and unexpected complications following successful FLP, that is, abruption, chorioamnionitis, fetal demise and that of a planned elective prematurity at 34–35 weeks following lung maturation by corticosteroids administration.12,56
Neonatal assessment of the hemodynamic and hematological status is a crucial component of the postnatal management. Cardiopathies because of intra-uterine cardiac remodeling are more evident after birth and can affect the survival of the neonate. Pregnancies with persistent vascular connections may result in a pair of anemic/polycythemic neonates, which may need immediate blood transfusion or exchange transfusion respectively.
Special case: TTTS in monoamniotic pregnancies
The monochorionicity of monoamniotic pregnancies predisposes it to TTTS. Ten percent of monoamniotic pregnancies are affected by TTTS.30 Instead of the typical oligo-poly sequence, only polyhydramnios is seen in the single shared amniotic cavity. All the other diagnostic markers can be evaluated like diamniotic pregnancies.
FLP can be performed for a monoamniotic pregnancy, but the proximity of cord insertions in often makes it difficult. Evidently, these fetuses also are at a high risk of cord entanglement.30 Hence, in cases where complete dichorionization is impossible to achieve, the practical option is selective feticide of either fetus by fetoscopic laser cord coagulation. This is followed by cord transection using a laser fiber under direct fetoscopic control. This ensures the de-tethering of the reduced fetus so that cord entanglement can be avoided.
Long-term outcomes following FLP
Survival and short-term outcomes are the principal factors influencing the line of management of TTTS cases. However, long-term outcomes are equally important for the expectant couple. The severity of the disease is directly and independently associated with neurological impairment. Higher Quintero stage at the time of diagnosis may lead to increased perinatal mortality and long-term morbidity.57,58
The findings of a large prospective cohort show that the FLP is associated with around 40% reduction in the risk of acute fetal death or long-term major neurological impairment, compared to amniodrainage.26 TTTS-affected pregnancies which underwent FLP had higher survival and lower morbidity at 6 months of age, and beyond that age the treatment itself has no major impact on long-term neurological outcome.7
Yves Ville conceived the idea and contributed to writing the manuscript. Viral M. Pandya wrote the manuscript. Claire Colmant and Julien Stirnemann contributed to writing the manuscript.5,6
Conflicts of Interest
. Quarello E, Ville Y. The feto–fetal transfusion syndrome story, an historical perspective. In: Twin-Twin Transfusion Syndrome. CRC Press;2007:1–8.
. Bajoria R, Wigglesworth J, Fisk NM. Angioarchitecture of monochorionic placentas in relation to the twin-twin transfusion syndrome. Am J Obstet Gynecol 1995;172(3):856–863. doi:10.1016/0002-9378(95)90011-x.
. Yamamoto M, Ville Y. Evidence on physiopathology. In: Twin-Twin Transfusion Syndrome. CRC Press;2007:29–40.
. Lewi L, Jani J, Cannie M, et al. Intertwin anastomoses in monochorionic placentas after fetoscopic laser coagulation for twin-to-twin transfusion syndrome: is there more than meets the eye? Am J Obstet Gynecol 2006;194(3):790–795. doi:10.1016/j.ajog.2005.08.062.
. Mahieu-Caputo D, Meulemans A, Martinovic J, et al. Paradoxic activation of the renin-angiotensin system in twin-twin transfusion syndrome: an explanation for cardiovascular disturbances in the recipient. Pediatr Res 2005;58(4):685–688. doi:10.1203/01.PDR.0000180558.03164.E8.
. Wieacker P, Wilhelm C, Prömpeler H, et al. Pathophysiology of polyhydramnios in twin transfusion syndrome. Fetal Diagn Ther 1992;7(2):87–92. doi:10.1159/000263655.
. Senat M-V, Deprest J, Boulvain M, et al. Endoscopic laser surgery versus serial amnioreduction for severe twin-to-twin transfusion syndrome. N Engl J Med 2004;351(2):136–144. doi:10.1056/NEJMoa032597.
. Hall JG. Twinning. Lancet 2003;362(9385):735–743. doi:10.1016/S0140-6736(03)14237-7.
. Quintero RA, Mueller OT, Martínez JM, et al. Twin–twin transfusion syndrome in a dizygotic monochorionic–diamniotic twin pregnancy. J Matern Fetal Neonatal Med 2003;14(4):279–281. doi:10.1080/jmf.14.4.279.281.
. Souter VL, Kapur RP, Nyholt DR, et al. A report of dizygous monochorionic twins. N Engl J Med 2003;349(2):154–158. doi:10.1056/NEJMoa030050.
. Salomon LJ, Ville Y. Le Syndrome Transfuseur-Transfuse: Prise En Charge et Traitement. 1st ed.Paris: un3fs; 2014.
. Khalil A, Rodgers M, Baschat A, et al. ISUOG practice guidelines: role of ultrasound in twin pregnancy. Ultrasound Obstet Gynecol 2016;47(2):247–263. doi:10.1002/uog.15821.
. Quintero R. Ultrasound assessment in twin–twin transfusion syndrome. In: Twin-Twin Transfusion Syndrome. CRC Press; 2007:71–90.
. Quintero RA, Morales WJ, Allen MH, et al. Staging of twin-twin transfusion syndrome. J Perinatol 1999;19(8):550–555. doi:10.1038/sj.jp.7200292.
. Rychik J, Tian Z, Bebbington M, et al. The twin-twin transfusion syndrome: spectrum of cardiovascular abnormality and development of a cardiovascular score to assess severity of disease. Am J Obstet Gynecol 2007;197(4):392e1–392e8. doi:10.1016/j.ajog.2007.06.055.
. Villa CR, Habli M, Votava-Smith JK, et al. Assessment of fetal cardiomyopathy in early-stage twin-twin transfusion syndrome: comparison between commonly reported cardiovascular assessment scores. Ultrasound Obstet Gynecol 2014;43(6):646–651. doi:10.1002/uog.13231.
. Stirnemann JJ, Mougeot M, Proulx F, et al. Profiling fetal cardiac function in twin–twin transfusion syndrome. Ultrasound Obstet Gynecol 2010;35(1):19–27. doi:10.1002/uog.7488.
. Stirnemann JJ, Nasr B, Proulx F, et al. Evaluation of the CHOP cardiovascular score as a prognostic predictor of outcome in twin-twin transfusion syndrome after laser coagulation of placental vessels in a prospective cohort. Ultrasound Obstet Gynecol 2010;36(1):52–57. doi:10.1002/uog.7713.
. Fisk NM, Tannirandorn Y, Nicolini U, et al. Amniotic pressure in disorders of amniotic fluid volume. Obstet Gynecol 1990;76(2):210–214.
. Fisk NM, Vaughan J, Talbert D. Impaired fetal blood gas status in polyhydramnios and its relation to raised amniotic pressure. Fetal Diagn Ther 1994;9(1):7–13. doi:10.1159/000263900.
. Bower SJ, Flack NJ, Sepulveda W, et al. Uterine artery blood flow response to correction of amniotic fluid volume. Am J Obstet Gynecol 1995;173(2):502–507. doi:10.1016/0002-9378(95)90273-2.
. Mari G, Roberts A, Detti L, et al. Perinatal morbidity and mortality rates in severe twin-twin transfusion syndrome: results of the International Amnioreduction Registry. Am J Obstet Gynecol 2001;185(3):708–715. doi:10.1067/mob.2001.117188.
. Rossi AC, D’Addario V. Laser therapy and serial amnioreduction as treatment for twin-twin transfusion syndrome: a meta analysis and review of literature. Am J Obstet Gynecol 2008;198(2):147–152. doi:10.1016/j.ajog.2007.09.043.
. Yamamoto M, Murr LE, Robyr R, et al. Incidence and impact of perioperative complications in 175 fetoscopy
-guided laser coagulations of chorionic plate anastomoses in fetofetal transfusion
syndrome before 26 weeks of gestation. Am J Obstet Gynecol 2005;193(3):1110–1116. doi:10.1016/j.ajog.2005.07.003.
. Quintero R. Umbilical cord occlusion in twin–twin transfusion syndrome. In: Twin-Twin Transfusion Syndrome. CRC Press;2007:121–126.
. Salomon LJ, Örtqvist L, Aegerter P, et al. Long-term developmental follow-up of infants who participated in a randomized clinical trial of amniocentesis vs laser photocoagulation
for the treatment of twin-to-twin transfusion syndrome. Am J Obstet Gynecol 2010;203(5):444e1–444e7. doi:10.1016/j.ajog.2010.08.054.
. Bebbington M. Twin-to-twin transfusion syndrome: current understanding of pathophysiology, in-utero therapy and impact for future development. Semin Fetal Neonatal Med 2010;15(1):15–20. doi:10.1016/j.siny.2009.05.001.
. Kontopoulos E, Quintero R. Treatment of twin–twin transfusion syndrome: an evidence-based analysis. In: Twin-Twin Transfusion Syndrome. CRC Press;2007:127–137.
. Chalouhi GE, Essaoui M, Stirnemann J, et al. Laser therapy for twin-to-twin transfusion syndrome (TTTS). Prenat Diagn 2011;31(7):637–646. doi:10.1002/pd.2803.
. Quintero R. Laser treatment for twin–twin transfusion syndrome. In: Twin-Twin Transfusion Syndrome. CRC Press; 2007:99–120.
. Bornick P, Allen M. Organization of nursing responsibilities. In: Twin-Twin Transfusion Syndrome. CRC Press;2007:159–166.
. Rossi A, Kaufman M. Anesthesia for surgical treatment of twin–twin transfusion syndrome. In: Twin-Twin Transfusion Syndrome. CRC Press;2007:183–194.
. Blizzard D. A trying experience: fetoscopy
and maternal decision making. Clin Obstet Gynecol 2005;48(3):562–573. doi:10.1097/01.grf.0000169658.73074.2c.
. Deprest J. Eurofoetus Group. Endoscopy in Fetal Medicine. Tuttlingen: Endo-Press; 2015.
. Slaghekke F, Lopriore E, Lewi L, et al. Fetoscopic laser coagulation of the vascular equator versus selective coagulation for twin-to-twin transfusion syndrome: an open-label randomised controlled trial. Lancet 2014;383(9935):2144–2151. doi:10.1016/S0140-6736(13)62419-8.
. Grannum PA, Copel JA. Invasive fetal procedures. Radiol Clin North Am 1990;28(1):217–226.
. Deprest JA, Van Ballaer PP, Evrard VA, et al. Experience with fetoscopic cord ligation. Eur J Obstet Gynecol Reprod Biol 1998;81(2):157–164. doi:10.1016/S0301-2115(98)00181-x.
. Quintero RA. Treatment of previable premature ruptured membranes. Clin Perinatol 2003;30(3):573–589. doi:10.1016/S0095-5108(03)00050-2.
. Morales W. Outcomes and complications of the surgical treatment of twin–twin transfusion syndrome. In: Twin-Twin Transfusion Syndrome. CRC Press;2007:145–149.
. Devlieger R, Millar LK, Bryant-Greenwood G, et al. Fetal membrane healing after spontaneous and iatrogenic membrane rupture: a review of current evidence. Am J Obstet Gynecol 2006;195(6):1512–1520. doi:10.1016/j.ajog.2006.01.074.
. Deprest J, Emonds M-P, Richter J, et al. Amniopatch for iatrogenic rupture of the fetal membranes. Prenat Diagn 2011;31(7):661–666. doi:10.1002/pd.2780.
. Baschat AA, Barber J, Pedersen N, et al. Outcome after fetoscopic selective laser ablation of placental anastomoses vs equatorial laser dichorionization for the treatment of twin-to-twin transfusion syndrome. Am J Obstet Gynecol 2013;209(3):234e1–234e8. doi:10.1016/j.ajog.2013.05.034.
. Stirnemann JJ, Nasr B, Quarello E, et al. A definition of selectivity in laser coagulation of chorionic plate anastomoses in twin-to-twin transfusion syndrome and its relationship to perinatal outcome. Am J Obstet Gynecol 2008;198(1):62e1–62e6. doi:10.1016/j.ajog.2007.06.009.
. Chalouhi GE, Stirnemann JJ, Salomon LJ, et al. Specific complications of monochorionic twin
pregnancies: twin–twin transfusion syndrome and twin reversed arterial perfusion sequence. Semin Fetal Neonatal Med 2010;15(6):349–356. doi:10.1016/j.siny.2010.09.003.
. Robyr R, Lewi L, Salomon LJ, et al. Prevalence and management of late fetal complications following successful selective laser coagulation of chorionic plate anastomoses in twin-to-twin transfusion syndrome. Am J Obstet Gynecol 2006;194(3):796–803. doi:10.1016/j.ajog.2005.08.069.
. Robyr R, Lewi L, Yamamoto M, et al. Permanent feto-fetal transfusion from the recipient to the donor twin. A complication of laser surgery in twin-to-twin transfusion syndrome. Am J Obstet Gynecol 2004;191(6):S163. doi:10.1016/j.ajog.2004.10.476.
. Slaghekke F, Kist WJ, Oepkes D, et al. Twin anemia-pPolycythemia sequence: diagnostic criteria, classification, perinatal management and outcome. Fetal Diagn Ther 2010;27(4):181–190. doi:10.1159/000304512.
. Lopriore E, Slaghekke F, Oepkes D, et al. Hematological characteristics in neonates with twin anemia-polycythemia sequence (TAPS). Prenat Diagn 2010;30(3):251–255. doi:10.1002/pd.2453.
. Quarello E, Molho M, Ville Y. Incidence, mechanisms, and patterns of fetal cerebral lesions in twin-to-twin transfusion syndrome. J Matern Fetal Neonatal Med 2007;20(8):589–597. doi:10.1080/14767050701449638.
. Hillman SC, Morris RK, Kilby MD. Co-twin prognosis after single fetal death: a systematic review and meta-analysis. Obstet Gynecol 2011;118(4):928–940. doi:10.1097/AOG.0b013e31822f129d.
. Shek NWM, Hillman SC, Kilby MD. Single-twin demise: pregnancy outcome. Best Pract Res Clin Obstet Gynaecol 2014;28(2):249–263. doi:10.1016/j.bpobgyn.2013.11.003.
. Weisz B, Hoffmann C, Ben-Baruch S, et al. Early detection by diffusion-weighted sequence magnetic resonance imaging of severe brain lesions after fetoscopic laser coagulation for twin-twin transfusion syndrome: fetoscopic laser coagulation in TTTS. Ultrasound Obstet Gynecol 2014;44(1):44–49. doi:10.1002/uog.13283.
. Lopriore E, Lewi L, Oepkes D, et al. In utero acquired limb ischemia in monochorionic twins with and without twin-to-twin transfusion syndrome. Prenat Diagn 2008;28(9):800–804. doi:10.1002/pd.2000.
. Winer N, Salomon LJ, Essaoui M, et al. Pseudoamniotic band syndrome: a rare complication of monochorionic twins with fetofetal transfusion
syndrome treated by laser coagulation. Am J Obstet Gynecol 2008;198(4):393e1–393e5. doi:10.1016/j.ajog.2007.09.030.
. Quarello E, Stirnemann J, Nassar M, et al. Outcome of anaemic monochorionic single survivors following early intrauterine rescue transfusion in cases of feto-fetal transfusion syndrome. BJOG Int J Obstet Gynaecol 2008;115(5):595–601. doi:10.1111/j.1471-0528.2007.01659.x.
. Stirnemann JJ, Quibel T, Essaoui M, et al. Timing of delivery following selective laser photocoagulation
for twin-to-twin transfusion syndrome. Am J Obstet Gynecol 2012;207(2):127e1–127e6. doi:10.1016/j.ajog.2012.06.042.
. Lopriore E, Middeldorp JM, Sueters M, et al. Long-term neurodevelopmental outcome in twin-to-twin transfusion syndrome treated with fetoscopic laser surgery. Am J Obstet Gynecol 2007;196(3):231e1–231e4. doi:10.1016/j.ajog.2006.10.906.
. Lopriore E, Nagel HTC, Vandenbussche FPHA, et al. Long-term neurodevelopmental outcome in twin-to-twin transfusion syndrome. Am J Obstet Gynecol 2003;189(5):1314–1319. doi:10.1067/S0002-9378(03)00760-9.