Chapter 27: FETAL THERAPY: MATERNAL FETAL SURGERY AND PERCUTANEOUS ULTRASOUND GUIDED FETAL THERAPY TECHNIQUES FOR CONGENITAL ANOMALIES

The field of fetal therapy is extensive and, other than for very specific experimental techniques, requires a multidisciplinary approach. Expertise from conception to the neonatal intensive care unit is necessary to move the endeavor forward and change the expectation issues from fetal survival to neonatal morbidity. The introduction of real-time ultrasound imaging into the practice of obstetrics in the early 1980s opened the maternal/fetal/uterine unit to a variety of prenatal diagnosis techniques with subsequent introduction of experimental therapies to treat congenital anomalies and genetic disease. While there are randomized controlled trials (RCTs) evaluating prenatal diagnosis techniques (amniocentesis, early amniocentesis, chorionic villus sampling), the same RCT evaluation for fetal therapy is limited. RCTs have been used in fetal therapy to evaluate monochorionic diamniotic twin pathology, specifically twin-to-twin transfusion syndrome, comparing amnioreduction with septostomy¹ and amnioreduction with laser vascular separation.^2,3 The majority of fetal therapy is evaluated as cohort comparisons and by expert opinion only. The development of "evidence-based outcome analysis" for fetal therapy is still required, and in most of this chapter the comments and opinions are expert opinions only. A "quality of evidence" and "strength of recommendation" is determined for each section (Table 27-1).⁴

This chapter is separated into 2 benefit/risk sections as follows:

Section I: maternal–fetal surgery (emergency surgery for fetal compromise/survival or elective surgery to improve neonatal morbidity) with maternal laparotomy and uterine hysterotomy with ultrasound contributions for hysterotomy site location and entry, and intraoperative ultrasound for fetal cardiac physiology/surveillance (Figure 27-1). Section II: percutaneous or "closed" techniques with ultrasound guidance and fetoscopic visualization for diagnosis, surveillance, localization, and treatment of the fetus with a proposed decrease in maternal morbidity/risk.

In the previous edition of this book, the chapter Invasive and Other Therapies was limited to fetal uropathy, obstructive hydrocephalus, and isolated pleural effusions. This new edition provides an expanded and detailed review of new topics and an updated summary of the fetal therapy topics for the reader's consideration with reference support. This chapter is restricted to the surgical fetal therapy approach. Ultrasound-guided correction for conditions such as fetal anemia, cardiac arrhythmia, or endocrine deficiency is not discussed.

The treatment of the fetus as a surgical patient has been clinically available as an "experimental therapy" since the early 1990s after extensive "proof of principle" surgical experiments with lamb and monkey research was completed.⁵ Harrison reviewed the rationale for fetal treatment considering selection, feasibility, and risk.⁶ The congenital anomalies detected in utero, but best corrected after delivery at term, are listed as gastrointestinal atresias, meconium ileus, enteric duplications/cyst, small intact omphalocele, small intact meningocele, unilateral hydronephrosis, craniofacial/ extremity/chest wall anomalies, cystic hygroma, small congenital tumors, and benign cysts (ovarian, mesenteric, choledochal). Other congenital anomalies that may benefit from correction at delivery using placental bypass with maternal "deep" general anesthesia (EXIT; ex utero intrapartum treatment) would include neonatal airway obstruction by cervical teratoma, cystic hygroma or congenital high airway obstruction (CHAOS), and lung masses preventing adequate ventilation (cystic adenomatoid malformation, bronchopulmonary sequestration, mediastinal teratoma, bronchial atresia). Debate continues about congenital anomalies that may benefit from "early" delivery to allow correction, but is limited to a "case-by-case" basis and balanced by the neonatal prematurity morbidity. This practice has not been widely supported, but a list of suggested anomalies would include urinary tract obstruction, hydrocephalus, gastroschisis, ruptured omphalocele, hydrops fetalis, intrauterine growth restriction, and certain cardiac arrhythmias.

Congenital anomalies considered for maternal–fetal surgery are divided into 2 major categories—fetal "life-threatening" defects and fetal "nonlethal" defects—with the purpose to reduce neonatal/childhood morbidity. Fetal life-threatening congenital defects have been identified after evaluating the natural history of certain congenital anomalies (without treatment) and then evaluating the potential risk/benefit for altering the natural history with surgical intervention. The congenital anomalies with an indication for maternal fetal therapy to date include (1) fetal hydrops (present or physiologically anticipated) secondary to congenital cystic adenomatoid malformation (CCAM), bronchopulmonary sequestration (BPS), and sacrococcygeal teratoma (SCT) (type I); and (2) nonhydropic congenital diaphragmatic hernia. At present, the only nonlethal congenital anomaly undergoing maternal–fetal surgery is myelomeningocele (MMC) (as part of a randomized research trial comparing the "gold" standard postnatal MMC repair with the "experimental" in utero MMC repair).

Open fetal surgical techniques are summarized in other publications and are not detailed in this chapter.⁵ Published outcomes are summarized in this chapter to enhance the "evidence-based" approach to counseling families about benefit/risk for the fetus and risk/benefit for the mother when prenatal imaging identifies a specific congenital anomaly in an "at-risk" fetus.

The present indications for maternal fetal surgery will vary from center to center, but at the Children's Hospital of Philadelphia the indications are summarized in Table 27-2 (Figure 27-2).

Fetal benefit/risk is part of the informed choice counseling with the understanding of the "elective" surgical choice (myelomeningocele) having a minimal "in utero" lethal risk and the "emergency" surgical choice (fetal hydrops) having an almost 100% risk of in utero or neonatal death dependent on the fetal anomaly and gestational age less than 30 weeks. Congenital anomalies are the leading cause of infant death (20%) and the US infant mortality rate in 2002 was 7.0 per 1000 liveborn infants.⁷ Long-term follow-up study followed 45,000 children with birth defects matched to 45,000 children without birth defects.⁸ Those children with birth defects had a 13 times higher rate of mortality (relative risk [RR] 12.9, confidence interval [CI] = 12.1 to 13.7), even after 10 to 15 years of follow-up. Higher neonatal death rates were associated with lower gestational ages and low birth weights. The number of congenital anomalies in the neonate had a strong dose-response curve. The range of adjusted risk of death in malformed children less than 2500 g and 5-year survival are summarized in Table 27-3. The specific congenital anomaly has a significant impact on morbidity/mortality and is a significant discussion point preoperatively.

The high postnatal morbidity and mortality rates for MMC is part of the rationale for this specific birth defect being considered as a candidate for in utero repair.⁹ MMC does not have a high in utero lethality risk, and it is usually isolated as a primary anomaly with secondary disruptive anomalies, has a 3% risk of chromosomal etiology, is relatively common (1 per 250 to 1000 conceptions depending on genetic/ethnic factors), and animal models have shown improved postnatal outcomes following in utero repair. Children with MMC have an increased morbidity due to brain stem dysfunction (Arnold-Chiari II malformation), decreased IQ (68% >80 IQ), cerebral ventriculomegaly requiring ventricular-peritoneal (VP) shunting, and motor/sensory deficits resulting in dysfunction of bowel, bladder, sexual, and lower limb areas.

Experimental maternal fetal surgery for the repair of MMC has provided preliminary maternal and fetal data to support an RCT comparing the "gold standard" treatment of postnatal MMC repair with the "experimental" treatment of in utero MMC repair at gestational ages of 20 weeks and 0 days to 25 weeks and 6 days.^{10,11, and 12} The pilot data, prior to the initiation of the RCT, showed that children who had undergone fetal MMC closure had characteristic neurodevelopmental deficits that did not appear to be worsened by fetal surgery, and developmental outcomes may have been improved due to a decreased rate of VP shunting (46% fetal repair vs 74% postnatal repair).¹⁰ Possible negative outcomes for the in utero repair identified after birth was the incidence of epidermoid cysts (57%) and symptomatic cord tethering at the mean age of 17 months. Family expectations for the "experimental" in utero therapy was evaluated by questionnaire and reported that families with spina bifida children (with postnatal repair) would want outcomes such as a VP shunt rate of 12%, wheelchair use at 8%, and urinary incontinence rate at 5% (50th centile questionnaire response) if in utero MMC repair therapy was to be considered or justified over the standard postnatal MMC repair.¹³

The maternal risk/benefit counseling is a more difficult component of the "informed choice" discussion as there are more psycho/social issues that need to be part of the family consideration. Maternal/paternal age, reproductive history, other children, family values, religious issues, neonatal morbidity/mortality issues, health insurance, geography, and health care facilities for ongoing neonatal/childhood care are some of the issues that must be considered during the counseling.

Maternal surgical/obstetrical risks have been reported for women undergoing maternal-fetal surgery for both emergency and elective indications. In an elective MMC cohort,¹¹ chorion-amnion separation is more likely to occur with surgeries done after 23 weeks' gestation. Membrane separation posthysterotomy was identified in 34% and contributed to an increased risk of premature rupture of membranes (PROM), but did not result in increased delivery rates before 30 weeks. Delivery prior to 33 weeks in this MMC group was 28%, but with an extreme preterm delivery rate resulting in a 7% (3 of 43) risk for neonatal death in this elective surgery cohort. The risk for intrasurgical or intrauterine death for the hydropic fetus undergoing emergency surgery is estimated at 50% depending on the congenital anomaly, the extent of the hydrops/cardiac dysfunction, and placentomegaly. For the elective surgery cohort,¹¹ an elective cesarean section delivery at 36 to 37 weeks' gestation was possible in 43% with an additional 22% having preterm labor and delivery from 33 to 36 weeks. Short-term morbidities were increased in women that had undergone maternal-fetal therapy requiring a hysterotomy and included increased rates of cesarean sections, intensive care unit (ICU) stay, prolonged hospitalization, and blood transfusion.¹⁴ To date, there is very limited information regarding the risk of maternal death associated with fetal therapy, but it is important to be aware that maternal mortality overall is under-reported.¹⁵

Subsequent pregnancy risk is a very important area for discussion with families prior to maternal-fetal surgery.^12,16 Maternal fertility does not appear to be affected by multiple hysterotomy scars.¹⁶ Preterm labor rates are estimated at 25% with full-term deliveries at 60%. Obstetrical complications are increased significantly in subsequent pregnancies with increased rates of uterine dehiscence (12%), uterine rupture (6%), cesarean hysterectomy (3%), and blood transfusion (9%).¹² Overall uterine rupture risks are similar to the "classical" cesarean section with reported risks of 4% to 9%. Other possible obstetrical issues to consider, but better studied in the routine cesarean section only population, are adverse outcomes in a subsequent pregnancy for stillbirth (odds ratio [OR] >2.3), placenta accreta (OR 18.75, CI = 2.28 to 864.6) and placenta previa (OR 1.66), or antepartum hemorrhage (OR 1.23). A strong recommendation for conception/birth spacing is made to women undergoing maternal-fetal surgery with intervals of 18 to 24 months from the "at-risk" delivery to subsequent conception. A delivery strategy for the subsequent pregnancy is based on computer modeling with the index pregnancy being a "classical" cesarean section. The recommendation is for a repeat cesarean section delivery at 36 weeks' gestation with no fetal lung maturity requirement/amniocentesis as this provides a lower risk of severe adverse outcome and higher maternal "quality of life."¹⁷

There are still many maternal and fetal issues to be identified and examined. The ongoing RCT for MMC will provide an enormous amount of data for the maternal and fetal surgical issues as well as maternal and neonatal/ childhood outcomes. The present maternal-fetal surgery data indicates that there is a good likelihood of positive fetal/neonatal improvement in morbidity and mortality with the transfer of morbidity and limited risk of mortality to the maternal patient in subsequent high-risk pregnancies. Informed choice will be very important with maternal, paternal, and family issues being considered.

The EXIT delivery has some increased maternal risk as deep general anesthesia, for uterine relaxation, is required to minimize placental separation and prolong the time for fetal intervention on placental bypass (mean 65 minutes).¹⁸ Maternal risks are placental abruption with increased maternal blood loss requiring blood transfusion. The most common hysterotomy location for the EXIT is the lower uterine segment, but placental and fetal position at times require hysterotomy sites in other uterine locations, which increases the present EXIT delivery risk as well as impacting subsequent delivery planning. Additional issues such as uterine atony, infection, and increased maternal pain and recovery time must be considered as part of the evidence-based discussion.¹⁹ The benefit to the fetus requiring an EXIT delivery is to minimize the anticipated neonatal morbidity, usually as airway or respiratory management. The EXIT delivery allows for airway management with neck tumors and resection of large fetal lung lesions, thereby avoiding acute respiratory decompensation related to mediastinal shift, cysts with air trapping, and compression of normal lung.¹⁸

Intraoperative fetal monitoring during maternal fetal surgery has shown its ability to provide real-time assessment of fetal hemodynamics, which allows intraoperative management changes with improved outcomes in the critically ill fetus. Intraoperative fetal echocardiography allowed identification of decreased ventricular filling, bradycardia, and decreased ventricular contractility such that administration of volume expanders and/or inotropic agents could be given to correct the abnormal physiology.²⁰ Neonatal survival was improved to 78% after implementation of intraoperative monitoring from 42% in the earlier maternal-fetal therapy cohort.²¹ The acute cardiovascular changes that take place during maternal-fetal surgery are likely a consequence of the physiology of the specific anomaly and the general effects of surgical stress, tocolytic agents, and general anesthesia.²⁰

Anesthesia considerations for fetal surgery have been reviewed.²² General anesthesia is the primary technique for hysterotomy-based surgical corrections of midgestation fetuses and EXIT delivery corrections of third trimester fetuses. Epidural analgesia, with general anesthesia as a possible backup, has been previously used for some fetoscopic cases, but more recently this type of percutaneous endoscopic therapy (twin-to-twin transfusion syndrome, bipolar cord occlusion, radiofrequency ablation) with ultrasound guidance components are done with maternal conscious sedation only and oral tocolysis (indomethacin, nifedipine).

Summary

The role of "open" maternal-fetal surgery for elective and emergency indications will require different types of evaluation for evidence-based outcomes. Ultrasound imaging is a required technology at all stages of the assessment, surgery, and follow-up of pregnancies complicated by significant fetal anomalies with increased risks for fetal morbidity and mortality. The emergency fetal therapy cases will never be evaluated as an RCT, and will require analysis of fetal and maternal morbidity as more important components than fetal survival only to support its validation and continued use. Elective fetal indications will require the completion of the RCT for MMC as this will allow careful and complete assessment of the maternal (index and future pregnancy), obstetrical, neonatal, and pediatric outcomes. Other "elective" fetal anomalies that could then be considered are gastroschisis, specific pathology with urinary obstruction, and possibly cerebral ventriculomegaly. The fetal risk involves early preterm delivery/ neonatal death estimated at 5%, and preterm delivery prior to 33 weeks' gestation at approximately 30%. Maternal morbidity will continue to be the major risk for maternal-fetal surgery with significant perinatal risk in subsequent pregnancies, but index pregnancy management continues to have important risks of preterm delivery and requires specific uterine evaluation and possible repair for dehiscence of the fundal hysterotomy scar, if present.

Fetal Thoracic Pathology (pleural effusion;macrocystic CCAM)

The specific techniques for the TA shunt procedure have been well documented and are not discussed in detail in this chapter.²³ In utero surgical intervention should be considered for fetuses with secondary complications such as early onset hydrops or progressive polyhydramnios. Initially, ultrasound-guided thoracocentesis should be utilized to drain the CCAM dominant macrocyst or pleural effusion, with serial ultrasound scanning to evaluate if the fluid reaccumulates, the time interval for reaccumulation, and to determine if chronic TA shunting will correct the physiology/ pathology. Polyhydramnios may require ultrasound-guided amnioreduction to keep the amniotic fluid volume within the normal gestational age limits. If polyhydramnios is present, maternal endovaginal ultrasound cervical measurements are necessary to assess the risks of preterm labor and delivery. Maternal oral broad-spectrum antibiotic prophylaxis is recommended following the TA shunt placement for 7 to 10 days. The prophylactic use of oral tocolysis (indomethacin; nifedipine) has not been evaluated in a systematic process and will vary from center to center. Tocolytic use will be gestational age dependent as well. The CHOP protocol uses oral indomethacin, 50 mg 1 hour prior to shunt insertion, and rarely requires any additional tocolysis (terbutaline 0.25 mg subcutaneously [SC] every 15 minutes (2 injections; nifedipine, 20 mg orally).

Fetal pleural effusions (PE) are divided into primary and secondary causes (Figure 27-3).²⁴ Primary PE is a lymphatic malformation while secondary PE has associated abnormalities such as cardiac dysfunction, anemia, infection, and aneuploidy. Primary PE has a frequency of 1 in 12,000 pregnancies with a 2:1 male-to-female ratio. Gestational age at presentation is usually less than 32 weeks' gestation, and the presence of hydrops is associated with high perinatal mortality rates (36% to 46%). The PE acts as a "space-occupying" lesion with resulting intrathoracic compression of the developing lungs and potential disturbance of intrathoracic blood flow secondary to the increased intrathoracic pressure and mediastinal shift of the heart and great vessels. The risk of lung deformation from the PE is directly related to the fluid volume. Bilateral PE has an increased risk for lung deformation/hypoplasia.

Indications for fetal therapy (TA shunt) for PE requires a primary lymphatic component PE etiology with normal karyotype, negative viral cultures, normal fetal echocardiogram, no other significant fetal anomalies, rapid reaccummulation of the PE after thoracocentesis (within 24 to 72 hours), and gestational age less than 32 weeks.²⁴ Perinatal outcomes have not been well evaluated for secondary PE etiologies treated with traditional fetal therapy techniques.

The most common complications of thoracic shunt procedures are displacement of the shunt into the amniotic cavity, or less commonly into the thoracic pleural space.²⁴ The shunt should be placed as low in the thoracic cavity as possible. Occlusion of the catheter is due to fetal blood collection at the insertion site or the viscosity of the proteinaceous material within the effusion. Pregnancy loss (preterm delivery) associated with shunt insertion is estimated at 5%. Thoracocentesis risks for fetal loss are estimated at 0.5% to 1.0%.

The benefit for the ultrasound-guided fetal therapy is dependent on patient/fetal selection. Aubard et al²⁵ identified fetal hydrops as the only prognostic feature for the outcome after multivariant analysis. Survival rate after TA shunt with or without hydrops was 67% and 100%, respectively, whereas without shunt treatment the survival was 21% to 23% in both groups.

Isolated fetal hydrothorax with hydrops was evaluated by systematic review of prenatal treatment options. Treatment options reviewed included thoracocentesis (single, serial), TA shunt, or combination. Overall population survival rate was 63%, ranging from 54% for single thoracocentesis to 67% with multiple thoracocentesis. In a limited cohort of 5 fetal cases treated with pleurodesis, there was an 80% survival. Shunt placement in fetuses with hydrops, with or without prior thoracocentesis, reported survival rates of 67% and 61%, respectively.²⁶ A summary of pleural effusions treated by TA shunting with minimal case reporting overlap had a total of 154 cases with hydrops in 64% (99) and a survival of 67% (103). Another "overlapping" comparison summary reviewed 203 cases (153 hydrops, 75%) reporting a survival with thoracocentesis only (35) of 60% and with TA shunting (158) of 66%, while the presence of hydrops decreased the survival rate to 50% and 62%, respectively.²⁷ An additional series of 54 hydropic fetuses treated with TA shunting had a survival rate of 57% but reported on the other TA shunt morbidities. Chorioamnionitis was found in 10% of the survivor group compared to 14% in the neonatal death group. Premature rupture of membranes occurred in 16% of the survivor group and 40% of the neonatal death group. The difference in shunt duration was 28 days for the survivor group and 7 days for the neonatal death group. The overall survival for fetuses with PE and hydrops based on these combined but overlapping summaries can be estimated at 65%.

Rustico et al²⁷ provide an algorithm for management of apparently isolated fetal PE, using 34 weeks' gestation as a significant triage gestational age. For fetuses less than 34 weeks, with hydrops and/or polyhydramnios, TA shunting is recommended, whereas fetuses with no hydrops and normal amniotic fluid would continue with weekly observation. For those fetuses greater than 34 weeks' gestation, consider thoracocentesis immediately prior to delivery.

One of difficulties with determining whether PE is isolated is congenital pulmonary lymphangiectasis (CPL), a primary lung lymphatic anomaly, which can only be identified by lung biopsy.²⁸ This rare CPL lesion increases the risk of hydrops (intestinal lymphatic component) being present but with a high neonatal lethality outcome.

Congenital cystic adenomatoid malformations (CCAMs) are benign space-occupying tumors caused by overgrowth of terminal respiratory bronchioles.^24,29 CCAMs are most often unilobar (80% to 95%) and may affect any lobe. They are usually classified by prenatal imaging (ultrasound [US], magnetic resonance imaging [MRI]) as microcystic (50%) or macrocystic (50%), depending on the size of the CCAM cysts. The CCAMs can occasionally contain a single or several dominant macrocysts, which are filled with lung fluid and can progressively enlarge until forming large space-occupying lesions within the fetal lung (Figure 27-4). The larger volume lesions can cause mediastinal shift, fetal hemodynamic compromise, and hydrops. Depending on gestational age when the large volumes are identified, critical lung development occurring between 18 and 24 weeks' gestation may be affected. Polyhydramnios can develop due to esophageal occlusion secondary to an enlarging lesion, with an increased risk of preterm labor. The rate of CCAM growth is increased between 20 and 25 weeks with an anticipated plateau of growth at 26 weeks' gestation. The natural history of CCAMs with a dominant macrocyst is more unpredictable than microcystic lesions due to the differential growth of the macrocystic and microcystic CCAM components.

Ultrasound surveillance, CCAM volume calculation, and hydrops risk assessment uses the cystic adenomatoid malformation (CAM) ratio or CCAM to head circumference ratio (CVR) for triage of the prenatally diagnosed fetus with a CCAM.²⁹ The CAM volume is obtained from the CAM ultrasound images with height, length, and width measurements and calculating the volume using the ellipse shape formula (length × height × width × 0.52). The CVR is then calculated by taking the CAM volume divided by the head circumference to correct for the gestational age. The CVR allows initial and serial evaluation of CCAM size or growth for risk assessment related to the development of hydrops. A CVR of greater than 1.6 at presentation predicts an increased risk of hydrops, developing due to the CCAM size. A CVR less than 1.6 at presentation suggests that the risk of hydrops developing in the absence of a dominant macrocyst is less than 3%.²⁹ Proposed fetal surveillance depending on the CVR at presentation is weekly if the CVR is less than 1.0; twice a week if the CVR is between 1.0 and 1.4; 3 times a week if the CVR is greater than 1.4. Fetal therapy options will vary dependent on the ultrasound appearance of microcystic or macrocystic CCAM lesions.

Medical therapy for CCAM therapy has been used in a small number of cases.³⁰ There is preliminary information that maternal steroid treatment (betamethasone 12 mg intramuscularly [IM] × 2 doses 24 hours apart) has an inhibiting effect on CCAM growth. Recent data evaluated 11 fetuses (5 with hydrops; 7 with CVR >1.6) with significant CCAM pathology and risk of hydrops development. CCAM growth rates were variable after maternal steroid administration, and when compared to historical controls, the CVR and CCAM growth rates decreased in 72% and 50% of patients, respectively, from the time of steroid administration to 28 weeks' gestation. This therapy has not been proven and a blinded/placebo RCT is planned to document the potential benefit for this noninvasive fetal treatment.

Because the fetal risks with the higher CVR macrocystic lesions, the TA shunt therapy may need to be used at earlier gestational ages. A case series reports that TA shunting at less than 21 weeks' gestation may increase the risk of a postnatal chest deformity due to rib development disruption.³¹ This potential risk should be included in the pre-procedure counseling with gestational ages less than 21 weeks at TA shunt insertion.

Laser therapy for CCAM, CCAM hybrid, and BPS pathology has been reported.^{32,33,34,35,36, and 37} Six cases have been reported with survival in 50%. Gestational age at treatment was 19 to 24 weeks. Table 27-4 summarizes the published reports.

The outcomes of fetuses with large macrocystic CCAMs treated with TA shunts are summarized in Table 27-5.^{24,38,39,40,41,42,43,44,45, and 46}

Other fetal thoracic pathology that may require in utero therapy includes mainstem bronchial atresia⁴⁷ and congenital lobar emphysema (CLE).⁴⁸ For the mainstem bronchial atresia, maternal-fetal surgery was unsuccessful with an intraoperative death at 24 weeks' gestation. CLE can have a prenatal ultrasound appearance similar to CCAM.

Evidence-based prenatal therapy for congenital thoracic anomalies (excluding congenital diaphragmatic hernia) is presented in Table 27-6.⁴⁹

Lower Urinary Tract Obstruction (LUTO)

Lower urinary tract obstruction is a heterogeneous fetal pathology group with posterior urethral valves and urethral atresia as the most common obstructive etiologies, accounting for one-third of the renal tract anomalies detected at autopsy after termination for ultrasound-diagnosed fetal anomalies.⁵⁰ Untreated fetuses with complete LUTO have abnormal renal development and function, leading to oligohydramnios, pulmonary hypoplasia, deformation type limb/facial anomalies with a high perinatal mortality rate. Although the posterior urethral valves and urethral atresia are specific for male fetuses, there are nonobstructive megacystis cases identified in female fetuses with a possible diagnosis of an autosomal recessive syndrome identified as megacystis, microcolon, hyperperistalsis syndrome.

Early fetal therapy procedures were directed at these urinary obstructive anomalies with bladder drainage by vesicocentesis and continuous drainage with vesicoamniotic shunt placement/drainage by vesicostomy using maternal-fetal surgery (Figure 27-5). These therapies have been available and used for many years due to the easily identified bladder obstruction and oligohydramnios appearance on ultrasound scanning. Systematic review/meta-analysis have shown that there is a lack of high-quality evidence to reliably direct clinical practice regarding prenatal bladder drainage in fetuses with ultrasonic evidence of LUTO. This lack of good evidence-based information emphasizes the need for protocol approaches to these obstructive bladder cases so that appropriate analysis can direct appropriate clinical decisions (treatment, no treatment).

The comparison of prenatal ultrasound imaging and postnatal anatomical pathologic analysis (autopsy) of male fetuses with LUTO prior to 25 weeks' gestation has shown that hyperechogenic kidneys were predictive of renal dysplasia in 95% of cases.⁵¹ In the 24 male cases, postmortem examination identified the pathological etiologies as urethral atresia (6), severe urethral stenosis (8), posterior urethral valves (9) and normal (1). Renal dysplasia was confirmed in 23 of 24 cases. Urethral atresia was the most common urethral anomaly at 12 to 17 weeks. Assessment of fetal bladder size and presence of hydronephrosis may possibly allow improved detection of the correct prenatal obstructive diagnosis. A saggital bladder diameter of 40 mm as associated with hydronephrosis before 28 weeks' gestation had a positive predictive value (PPV) for posterior urethral valves of 44%, the absence of hydronephrosis and bladder diameter of less than 40 mm had a PPV for urethral atresia/ stenosis of 100%, and the absence of hydronephrosis had PPV of urethral atresia of 67%.

Prenatal selection criteria for vesicoamniotic shunting are isolated megacystis, bilateral hydronephrosis, decreased amniotic fluid volume, no other associated congenital anomalies, normal male karyotype 46,XY, and serially improving urinary/renal function values (Table 27-7).²³

In serial ultrasound-guided vesicocentesis it is recommended that 3 fetal specimens be obtained over 5 to 7 days, showing improving renal function values but with specimen 2 or 3 reflecting retained renal function parameters, thereby identifying the male fetus as a candidate that may benefit from vesicoamniotic shunting (VAS).²³ The VAS insertion technique has been described.²³

Systemic review has compared the effect of bladder drainage versus no bladder drainage on perinatal survival.⁵² Among the controlled studies, bladder drainage appeared to improve perinatal survival relative to no drainage (OR 2.5; 95% CI = 1.1 to 5.9; P = .03). The contradictory finding from this review was that this improved outcome observation was largely in the subgroup of fetuses with a prenatally predicted poor prognosis based on ultrasound and fetal urinary electrolytes criteria (OR 8.1; 95% CI = 1.2 to 52.9; P = .03). An improved outcome was also suggested in the good prognosis group (OR 2.8; 95% CI = 0.7 to 10.8; P = .13) but without significant values. The conclusions from this systematic review reported that the limited available evidence suggests that prenatal bladder drainage may improve perinatal survival in these bladder-obstructed fetuses, particularly those fetuses with a poor predicted renal prognosis by ultrasound and urinary renal function parameters.⁵²

An RCT (PLUTO, University of Birmingham, UK) is presently enrolling fetuses to investigate the role of fetal vesicoamniotic shunting in moderate/severe LUTO.⁵³ The study plans to recruit 200 women over 3 years with the criteria of a singleton pregnancy, viable intrauterine gestation with a diagnosis of isolated bladder outflow obstruction at less than 28 weeks (hydronephrosis, megacystis, oligohydramnios less than the 5th centile, and no other structural or chromosomal abnormalities). Fetuses will be randomized to VAS or conservative observation following maternal consent. Follow-up is scheduled for 5 years looking at obstetrical outcomes, procedural complications, survival, renal function/pathology, disability, and continence.

Complications from the VAS must be part of the informed consent process. These risks include chorioamnionitis, premature rupture of membranes, direct trauma to the fetus, placental bleeding, and possible preterm labor. Transient vesicoperitoneal fistulas occasionally result in urinary ascites. Fistula closure usually takes 10 to 14 days. Shunt displacement occurs in 30% to 45% of VAS cases. The fetal death rate following shunt placement is reported to be 5% (9 of 169 cases). Vesicocentesis risks for pregnancy loss are considered similar to amniocentesis rates at 0.5% to 1.0%, but are additive with the serial procedure requirements.

VAS under ultrasound guidance is usually done at gestational ages of 20 to 30 weeks following identification and evaluation of the fetal urinary obstruction.²³ VAS has been done as early as 13⁺⁵ weeks using a double-basket catheter in a fetus with posterior urethral valves (PUV), delivery at 33 weeks, and normal renal function at age 3 years.⁵⁴

Large database evaluation has reported on the diagnosis and outcome of fetal lower urinary tract obstruction.⁵⁵ The incidence of LUTO was 2.2 per 10,000 births. The prenatal detection improved over the study period from 33% to 62%. The sensitivity for prenatal ultrasound prediction of renal dysplasia on autopsy or chronic renal failure based on fetal urine parameters (sodium, calcium or β₂-microglobulin) was 59%, 33%, 66%, and 63%, respectively. This study supports the poor prognosis natural history for fetuses with LUTO.

Long-term outcome for 23 pregnancies complicated by antenatal oligohydramnios had 9 male fetuses with posterior urethral valves and survival in 7 (78%).⁵⁶ Developmental delay was present in 3 of the 7 survivors. Amnioinfusion was used in 1 survivor at 19 weeks' gestation with postnatal outcome reporting neonatal peritoneal dialysis, vascular dialysis at age 2, and renal transplant at age 4. Postnatal renal function was variable, but long-term outcome results in the survivors was reported to be encouraging. Biard et al reported long-term outcomes in fetuses treated with in utero shunting and found significant morbidity in survivors.⁵⁷ Clinical outcomes from 20 singleton male fetuses with oligohydramnios and LUTO had a 1-year survival of 91% (with 2 neonatal deaths from pulmonary hypoplasia). Mean age to follow-up was 5.8 years. Pathology for the LUTO was PUV (7), UA (4), and prune belly syndrome (7).

Neurodevelopmental outcomes for this group were 78% normal with learning, speech, and physical therapy issues in 11%, 16%, and 16%, respectively. Eight children had acceptable renal function, 4 had mild renal insufficiency, and 6 required dialysis with eventual renal transplant. Other comorbidities included respiratory (8) and musculoskeletal problems (9). Self-completed Quality of Life questionnaires showed no reported difference compared to healthy controls in a validated evaluation for parents and children in this cohort.

Congenital megalourethra is characterized by ultrasound features of LUTO in association with dilatation and elongation of the penile urethra.⁵⁸ Severe oligohydramnios was found in only 1 of 4 prenatal cases between 20 and 24 weeks' gestation. Neonatal deaths occurred in 3 cases with 1 pregnancy termination. No prenatal therapy was used in these cases.

Urinary congenital anomalies for female fetuses involves cloacal anomalies (isolated) or as part of a larger OEIS complex (omphalocele, bladder extrophy, imperforate anus, spinal defect). Cloacal anomaly is a contraindication to in utero therapy, as isolated attempts to shunt large cystic spaces have not indicated benefit. OEIS has variable outcomes but survival with significant genitourinary/ reproductive morbidity is possible, even with no in utero therapy considerations.⁵⁹

Megacystis-microcolon-intestinal hypoperistalsis syndrome (MMIHS) presents most commonly in females with megacystis and normal amniotic fluid volume. Vesicocentesis has been reported with vesicoamniotic shunting.⁶⁰ Normal renal function is usually identified with urinary fluid analysis. The vesicocentesis to completely empty the bladder allowed the dilated fetal bowel to be identified, confirming the MMIHS diagnosis. Amniotic fluid analysis for fetal gastrointestinal enzymes has also been used in conjunction with the normal renal function findings to diagnose MMIHS.⁶¹ Survival with an MMIHS diagnosis is estimated at 70%.

Monochorionic Twin Pathology: Twin-to-Twin Transfusion Syndrome (TTTS)

The extensive literature regarding TTTS can only be summarized. Review articles summarize the topic (Figure 27-6).⁶² Monochorionic diamniotic twinning has a higher risk of perinatal complications than dichorionic twinning, even after exclusion of disorders unique to monochorionic placentation.⁶³ Severe TTTS occurs in 10% to 15% of monochorionic diamniotic twin pregnancies and carries a significant risk of perinatal morbidity and mortality. TTTS results from placental vascular anastomoses that result in a differential distribution of blood and vasoactive substances between the donor and recipient twin. The natural history of severe TTTS results in extreme preterm delivery (prior to 25 weeks) with neonatal death in 90%. Disease severity in TTTS is predicted by the ultrasound parameters in the Quintero staging system, which describes progressive changes that begin with intertwin differences in growth and amniotic fluid volume, and progress to an absence of visualization of the urinary bladder as a surrogate for hypovolemia in the donor, abnormalities of umbilical Doppler ultrasonography, and ultimately the development of hydrops and fetal death in the overloaded recipient.⁶⁴ Whereas the Quintero staging system has provided a valuable platform for comparison and staging of TTTS between pregnancies and institutions, fetal echocardiography evaluations have provided new insight and assessment into the pathophysiology of TTTS. Fetal cardiac parameters (ventricular hypertrophy, dilatation, function, valve regurgitation, great artery size, and diastolic properties in the recipient and umbilical artery flow in the donor) have been shown to enhance the clinical assessment for staging and treatment triage as well as to evaluate post-endoscopic laser treatment success.⁶⁵

Several treatment options have been used over a number of years with the goal of improving neonatal outcome. Techniques that have been used include serial amniodrainage, intertwin amniotic septostomy, selective termination of one fetus, and the present treatment, "gold standard" endoscopic laser coagulation of the placental vascular anastomosis.^{1,2, and 3}

Fetoscopic laser coagulation for TTTS was first described by Dr Julian DeLia as a technique that functionally converts a monochorionic placenta to a functionally dichorionic placenta by occluding vascular connections between the twins. A subsequent laser technique used a more selective location for the coagulation site, as the vascular equator between the twins did not correspond with the placental location of the intertwin membrane.^{1,2, and 3,64} This selective laser technique is hypothesized to preserve noncommunicating vessels and has been associated with improved survival rates of 62% to 77%.

Two RCTs have evaluated the TTTS treatment with comparison between the previous "gold standard" amnioreduction (AR) and the experimental endoscopic laser coagulation/separation.^2–3 These 2 studies have contradictory conclusions with the "Eurofetus" trial² corresponding with a larger body of cohort studies. The "Eurofetus" study identified pregnant woman with severe TTTS prior to 26 weeks' gestation and randomized 70 women to AR and 72 to endoscopic laser coagulation. The laser group had a higher likelihood of the survival of at least 1 twin to 28 days of age (76% vs 56%). Infants in the laser group had a lower incidence of neurological injury (6% vs 14%). Endoscopic laser coagulation of anastomosis was a more effective first-line treatment than serial AR for severe TTTS diagnosed before 26 weeks' gestation. The National Institute of Child Health and Human Development (NICHD) trial³ was not fully completed and the trial was closed to enrollment as the TTTS "Eurofetus" results caused clinicians to be convinced that endoscopic laser coagulation was the treatment of choice for severe TTTS.

Prospective studies continue to report improved neonatal outcomes with endoscopic laser coagulation. Huber et al reports an overall survival rate of 71.5% with survival of both twins at 59.5% and survival of at least 1 twin at 83.5% in a 200-twin-pair, consecutive pregnancies cohort.⁶⁶ The mean gestational age at delivery was 34.3 weeks (23 to 40 weeks). There was a reduced neonatal survival as the TTTS increased in staged severity: stage I—both twins 76%, at least one 93%; stage II—both twins 60%, at least one 83%; stage III—both twins 54%, at least one 83%; stage IV—both twins 50%, at least one 70%. The overall survival rate for donor fetuses was 70.5% and for recipient fetuses 72.5%. Other studies continue to support this level of outcome.^67,68

Long-term neonatal outcome studies report good neurological and growth outcomes with follow-up intervals of 2 to 6 years. Graef et al⁶⁹ report on 167 surviving infants from TTTS with endoscopic laser coagulation followed to a median age of 3 years and 2 months. Normal neurodevelopmental outcomes were seen in 86.8%, and an incidence of major neurological deficiencies in 6%. Lopriore et al⁷⁰ reports on 82 twin pairs from TTTS and a survival of 70% with neurodevelopmental impairment in 17% (19 of 115) including cerebral palsy (8), mental developmental delay (9), psychomotor developmental delay (12), and deafness (1). Lenclen et al⁷¹ assessed the neonatal outcome in 137 TTTS preterm neonates treated with AR (36) and laser (101). These 137 monochorionic twins were compared to 242 dichorionic twins. Gestational ages at delivery for the groups was 24 to 34 weeks. Adverse neonatal outcomes (death or severe cerebral lesions) was more frequent in the AR group. The overall neonatal outcomes were comparable for the laser and dichorionic groups. However, the neonatal morbidity was higher in the endoscopic-laser-exposed neonates born at 30 weeks' gestation mainly due to failed laser therapy. Lopriore et al⁷⁰ reported on monochorionic twins with TTTS treated with endoscopic laser compared to monochorionic twins without TTTS. Antenatally acquired severe cerebral lesions in the TTTS and non-TTTS groups was 10% and 2%, respectively. At discharge, the incidence of severe cerebral lesions in the TTTS and non-TTTS groups was 14% and 6%, respectively. Antenatal injury was responsible for severe cerebral lesions in 67%. A smaller non–endoscopic laser Japanese study of 18 monochorionic twins (33 survivors) and 12 treated with AR were followed to age 6.⁷² No difference in cerebral palsy, epilepsy, mental retardation, neurological deficits at school age, or growth when compared to weight-matched singleton controls was found. The median and range for diagnosis and birth for the TTTS pregnancies were 23 (10 to 33) and 28.5 (23 to 36) weeks, respectively.

Specific fetal cardiac features of TTTS were introduced in the screening evaluation portion of this chapter with the CHOP cardiovascular score for TTTS.⁶⁵ In TTTS, the donor and recipient are exposed to differing volume loads and show discordant intertwin vascular compliance in childhood despite identical genotype. Vascular programming is evident in monozygotic twins with intertwin transfusion and is altered but not abolished by intrauterine therapy to resemble that seen in dichorionic twins.⁷³ A similar analysis to evaluate cardiac morphology and function in survivors of severe TTTS after laser coagulation reported similar cardiac changes post-laser with improved biventricular systolic function and tends to improve diastolic function in the recipient twin.⁷⁴ Barrea et al reported on the fetal cardiovascular changes in TTTS recipients after AR. Despite AR, the recipient twin had progressive biventricular hypertrophy with predominant right ventricular systolic and biventricular diastolic dysfunction.⁷⁴ The prevalence of congenital heart disease (particularly pulmonary stenosis) was increased in the TTTS recipients with 11.2% and 7.8% compared to the general population with 0.3% and 0.03%, respectively.

Renal function in twins complicated with TTTS is affected by the hypertensive status in the recipient (polyuria) and hypotension status in the donor (anuria).⁷⁵ No significant difference in long-term renal function was found between recipient and donors from 18 surviving twin pairs after endoscopic laser coagulation. The laboratory findings for all 36 children were within normal limits at mean age 3 years and 1 month (range 1 year and 9 months to 4 years and 5 months).

Fetal growth analysis of monochorionic twins after endoscopic laser treatment for TTTS has shown a slowdown in recipient growth and maintained growth in the donor, leading to a decrease in intertwin discordance.

Maternal cervical length as a surrogate marker for the maternal risk of preterm labor was evaluated in a cohort of severe TTTS following endoscopic laser coagulation prior to 26 weeks' gestation. The mean cervical length was 32 and 38 mm in women delivering before and at or before 34 weeks, respectively (P < .0001). For a cervical length of less than 30 mm, the risk of delivery before 34 weeks was 74%. Three independent factors to predict preterm delivery were identified: short cervical length (increased risk), parity (increased risk), and intrauterine death of 1 twin (decreased risk).⁷⁶

Perioperative endoscopic laser complications occur, and this potential risk should be part of the informed consent counseling. Perioperative complications were reported from 175 percutaneous endoscopic laser procedures completed under local anesthesia.⁷⁷ Placental abruption and miscarriage occurred in 3 and 12 (7%) cases, respectively. Premature rupture of membranes (PROM) occurred in 28% (49 cases) with 12, 29, and 46 cases that occurred before 24, 28, and 34 weeks, respectively. The trocar entry tract was transplacental in 27% (48 cases) but no significant adverse outcomes resulted.

Late fetal complications following 151 consecutive successful laser therapy cases reported miscarriage, double intrauterine fetal death (IUFD), and single IUFD occurred in 4% (6 of 151), 5% (7 of 151), and 26% (39 of 151), respectively. At 7 days after the procedure, 67% (101 of 151) had 2 living fetuses. Recurrence of the TTTS features occurred in 14 cases (14%). Middle cerebral artery-peak systolic velocity (MCA-PSV) evaluation identified 2 cases with values indicating anemia (increased) and polycythemia (decreased). This result was hypothesized to be due to unidirectional fetofetal blood transfusion, mainly from former recipients into former donors.⁷⁸ Late complications were managed accordingly by repeat laser, AR, cord coagulation, intrauterine blood transfusion, or elective delivery. MCA-PSA Doppler measurements were useful in double survivor follow-up to identify late vascular changes.

When the diagnosis of monochorionic diamniotic twins with severe TTTS is made, therapeutic AR should be discouraged in patients considering endoscopic laser coagulation and should only be utilized for maternal findings of cervical dilation/internal cervical os "beaking" or maternal respiratory compromise. Membrane detachment from the uterine wall is a complication of AR and results in delayed or no access to endoscopic laser coagulation therapy. Amniopatch techniques have been proposed to treat this TTTS complication.⁶⁴

At the present time, monochorionic diamniotic twin pairs complicated by severe TTTS prior to 26 weeks have improved outcomes for survival, gestational age at delivery, and outcomes for neurological, renal, and cardiac disease if treated with endoscopic laser coagulation therapy. The experience of the operator is an important aspect in the successful treatment outcome. Treatment centers should attempt to maintain caseload volumes of greater than 40 cases per year for maintenance of skills and management by the operative team (medical, anesthesia, nursing).

Monochorionic Twin Pathology and Selective Termination for TTTS, TRAP, Discordant Fetal Anomaly

Monochorionic twinning is a congenital anomaly with the timing of the embryonic splitting determining the fate of the twin pair. The placental vascular sharing contributes to monochorionic twin pathologies including severe twin-to-twin transfusion syndrome (TTTS), twin reversed arterial perfusion (TRAP), discordant fetal anomalies, and placental insufficiency with discordant intrauterine fetal growth restriction (IUGR).

TTTS has been described in this chapter. TRAP or acardiac twinning has a prevalence of 1 in 35,000 pregnancies. The presence of an arterioarterial and venous anastomosis between the cord insertions in monochorionic twins may lead to a reversal of perfusion via the umbilical artery of one twin, if one pulse wave predominates over the other pulse wave, early in the gestation development (Figure 27-7).⁷⁹ In the reversely perfused twin, there is no cardiac development at all or only a rudimentary heart tube can be detected. The development of the upper part of the body is severely impaired, and most acardiac twins show acrania and hydrops.⁷⁹ The normally developed "pump" twin is at risk for heart failure, preterm delivery, and intrauterine death. The "pump" twin mortality was 51% and delivery prior to 34 weeks occurred in 76%.

A wide variety of innovative techniques have been reported for the selective separation of the normal "pump" twin from the acardiac twin. Quintero et al reported on 74 monochorionic twin pregnancies complicated by TRAP (1993 to 2004) with 51 pregnancies having a procedure for cord occlusion to the abnormal twin.⁸⁰ Because of the reported time period, a variety of techniques were used for umbilical cord occlusion (UCO) including umbilical cord ligation, UCO via laser coagulation of the cord, cord ligation and transaction (monoamniotic twin pair), and endoscopic laser coagulation of the arterioarterial and venovenous anastomoses. The overall perinatal survival in the surgical group was 65% compared to 43% in the nonsurgical group. Surgery within the acardiac twin's gestational sac was possible in 24%. There were no significant differences in the perinatal outcomes relative to the specific surgical technique reported. Hecher et al⁷⁹ reports on 60 consecutive pregnancies complicated with TRAP using endoscopic laser treatment of placental anastomoses (18) and umbilical cord (42). Median gestational age at treatment was 18.3 weeks (14.3 to 24.7 weeks). Vascular coagulation with arrest of blood flow was achieved in 82% by laser alone and in a further 15% by laser coagulation in combination with bipolar forceps. The overall survival of the "pump" twin was 80%. Median gestational age at delivery was 37 weeks (24 to 41 weeks). PROM prior to 34 weeks' gestation occurred in 18%. One of the proposed advantages reported by Hecher et al is that this technique can be used at gestational ages less than 18 weeks as bipolar cord cautery was reported to have an increased fetal demise when performed at 16 to 17 weeks compared to 18 weeks or later (41% vs 3%).

Lewi et al report on selective fetal termination by cord coagulation in 80 complicated monochorionic multiple pregnancies (67 monochorionic diamniotic [MCDA]; 6 monochorionic monoamniotic [MCMA]; 4 monochorionic triamniotic [MCTA]; 3 dichorionic triamniotic [DCTA]) using endoscopic laser and/or bipolar techniques.⁸¹ The indications for the fetal cord cautery were reported as discordant fetal anomaly (35%), TTTS (30%), TRAP (27.5%), and severe IUGR (7.5%). For 69% (55 of 80) of cases, laser was used as the primary technique, but half the cases required additional bipolar coagulation to achieve arrest of flow. Bipolar cord cautery was used as the primary technique in the remaining 31% (25 of 80). Median gestational age at procedure was lower in the primary laser group compared to the bipolar group (19.4 weeks vs 22.4 weeks). Only 1 bipolar case was unsuccessful in a TRAP with cord perforation and an intraoperative IUFD. The survival rate was 83% (72 of 87) with no significant differences for twins and triplets. There were 9 IUFDs (10%) with 5 early and 4 late losses. There was 1 pregnancy termination and 5 perinatal deaths due to PROM prior to 25 weeks' gestation. Ongoing pregnancies had a mean gestational age at delivery of 35 weeks. Delivery before 28 weeks gestation was 85 (6 of 71) and between 28 and 32 weeks in 13% (9 of 71). Preterm PROM (PPROM) occurred in 38% (27 of 71). Developmental outcome was available for all 72 survivors, and the children greater than 1 year of age 92% (62 of 67) had normal development.

Each of these complicated monochorionic and monochorionic multiple pregnancies are individual and unique due to the placental locations, twin pathology, maternal body mass index (BMI), gestational age at presentation, and a variety of the maternal/fetal/obstetrical issues. Techniques and approaches will be dependent on operator expertise and experience. The above reports have skilled operators using multiple and combined techniques for the selective termination. Tables 27-8^{82,83,84,85,86, and 87} and 27-9^{88,89, and 90} compare the ultrasound-guided percutaneous single device techniques of bipolar umbilical cord cautery (BUCC) and radiofrequency ablation (RFA) with the complicated monchorionic twin/triplet pathology for selective termination. While RFA is a purely ultrasound-guided technique with ultrasound Doppler assessment for completion of blood flow arrest, BUCC is an ultrasound-guided technique for the bipolar instrument port insertion and cord location/ cautery/cord transection. The fetoscope can be used via the 3-mm port to visualize the cord cautery coagulation sites, but confirmation of blood flow occlusion requires Doppler flow evaluation. The RFA technique does not require amnioinfusion of the abnormal gestational sac if oligohydramnios is present, while BUCC may, in certain situations, require this additional procedure of amnioinfusion to allow umbilical cord access.

The technique that provides the best benefit:risk for the selective termination of a monochorionic twin with discordant anomalies or physiology has not as yet been determined. It is not likely that an RCT will be available to determine the best technique for these indications, but gestational age and vascular diameter may determine that RFA can be used at gestational ages less than 20 weeks and bipolar cord cautery may be a better choice for gestational ages greater than 20 weeks. Post-procedure loss rates are estimated at 20% for RFA and 15% for bipolar cord cautery. PROM rates for both techniques are estimated at 18% to 20% (Table 27-10).

Other monochorionic placental functional abnormalities can result in unequal placental sharing with one twin having normal growth while the other twin has IUGR. In singleton pregnancies, fetuses with IUGR are at higher risk for poor perinatal and long-term outcome compared to fetuses with appropriate growth. Placental factors are the most significant etiology for growth restriction interrelated to cardiovascular and behavioral responses, and metabolic status. Factors that impact the neonatal outcomes in singleton fetuses with placental-based growth restriction are gestational age, birth weight, and ductus venosus Doppler parameters.⁹¹ Other physiological assessment has evaluated Doppler values (umbilical artery [UA], ductus venosus [DV]) and short-term fetal heart rate variation to predict perinatal outcome in the severe, early IUGR fetus. The DV pulsatility index for veins (PIV) measurement was the best predictor of perinatal outcome with a cutoff value of 2 to 3 SD as appropriate for delivery. Considering placental insufficiency perinatal outcomes in monochorionic twin pregnancies, amniotic fluid discordance, IUGR, and umbilical artery absent or reversed end-diastolic (ARED) flow in one fetus represents an extremely high-risk constellation for adverse pregnancy outcome.

Placental Chorioangioma Management With Ultrasound-Guided Techniques

Placental Chorioangioma Management With Ultrasound-Guided Techniques Chorioangiomas are placental lesions due to an abnormal proliferation of blood vessels in chorionic villi. Histologically, they consist of small blood vessels that are embedded within the stroma of enlarged placental villi and are covered with a trophoblastic layer. They are the most common placental tumor with a 1% incidence. Sepulveda et al followed 11 pregnancies with placental chorioangiomas.⁹² The mean age at diagnosis was 26 weeks (range 20 to 36 weeks). Size was greater than 4 cm in 9 of 11 cases. Polyhydramnios and preterm labor were the most common pregnancy complications. Amnioreduction at 27 weeks was used in one case with a successful term delivery. A second treatment case of alcohol ablation was used at 26 weeks in a hydropic/cardiac failure fetus with early post-treatment fetal demise. Ultrasound surveillance was used including fetal biometry, amniotic fluid volume, and Doppler flow studies, with fetal echocardiography in selected cases. Serial ultrasound assessments with nonstress fetal heart rate monitoring were used in the third trimester.

Guschmann et al examined 22,439 unselected third trimester placentas and identified chorioangiomas in 0.61% of the pregnancies.⁹³ The incidence of chorioangiomas corresponded with increasing maternal age, most often in women greater than 30 years of age, as well as maternal hypertension and diabetes. The rate of preterm birth was increased (3×) in the pregnancies complicated by chorioangiomas. A rare neonatal complication of chorioangiomas are diffuse, multiple hemangiomas.⁹⁴ There is a case report of a euploid female fetus with prenatal identification of a chorioangioma and delivery at 35 weeks due to polyhydramnios. Maternal serum screening was abnormal with an α-fetoprotein (AFP) level of 14.9 multiples of the median (MoM). The newborn was found at birth to have multiple diffuse cutaneous and liver angiomatosis. Neonatal death occurred at 1 month of age secondary to cardiac failure and infection.

Chorioangioma complications in the fetus/neonate, and maternal and placental compartments occur in the antepartum, intrapartum, and postpartum time periods.^{95,96,97,98,99,100, and 101} Most chorioangiomas are small and have no clinical significance, but lesions greater than 40 mm are associated with pregnancy complications. Fetal effects include antepartum cardiac failure, hepatosplenomegaly, IUGR, microangiopathic anemia, thrombocytopenia, fetomaternal hemorrhage with intrapartum nonreassuring fetal heart rate tracing, and newborn heart failure, apnea, and hemangiomas. Maternal complications are preterm labor (polyhydramnios, antepartum hemorrhage, PROM), pregnancy-induced hypertension, proteinuria, thrombocytopenia, and postpartum hemorrhage. Direct placental complications include abruption and infarction with placental retention at delivery.

Ultrasound-guided placental chorioangioma treatments have included laser therapy (interstitial tumor, surface feeding vessel), intrauterine transfusion, embolization (enbucrilate, microcoil), and alcohol injection. Timing and outcomes are summarized in Table 27-11.^{95,96,97,98,99,100, and 101}

Umbilical cord lesions are uncommon, but ultrasound will usually allow prenatal diagnosis. Hemangiomas of the umbilical cord are typically seen as fusiform swellings in the cord with an angiomatous nodule and edematous Wharton jelly. The hemagioma location is usually close to the placental cord insertion site. Ultrasound-guided decompression of the cystic portion of the hemangioma has been used in the third trimester to improve umbilical Doppler flow, and to prevent fetal malpresentation and dystocia, and minimize the risk of cyst rupture. This decompression management allowed a vaginal delivery to occur safely.

Fetal Ovarian Cyst

Ovarian cysts are a common abdominal mass in female fetuses with an estimated incidence of 1 in 2600 fetuses.^102,103 The cysts develop due to fetal follicle-stimulating hormone (FSH), maternal estrogens, and human placental chorionic gonadotropin (hCG) stimulation. Other obstetrical complications associated with increased hCG production (maternal diabetes, preeclampsia, rhesus sensitization) have an increased incidence of ovarian cysts. Ultrasound criteria for a probable ovarian cyst are female gender, fetal pelvic location, cystic appearance (simple, complex), and no peristalsis. Simple cysts are anechoic with no internal debris or septations. Complex cysts have internal echos, septations, and debris. Meta-analysis has identified 420 fetuses with ovarian cysts (1984 to 2005). In 50% of the cases, the cyst regressed spontaneously, while 35% of cysts were complicated by torsion and intracystic hemorrhage. Cysts less than 50 mm regressed spontaneously in 98%. Cysts greater than 50 mm resulted in a complicated course in 93% and direct recommendations to consider decompression of the cyst by ultrasound-guided needle aspiration to prevent ovarian torsion. Following the aspiration, spontaneous regression was observed in 89%. Prenatal aspiration appears to be effective and safe. Cysts with an ultrasound appearance of torsion that persist postnatally require surgical evaluation. The neonatal outcomes following a protocol of prenatal serial ultrasound without in utero cyst aspiration and early postnatal ultrasound scan are reported.¹⁰³ Neonatal surgery was indicated for complex-appearing cysts regardless of size, and for simple cysts greater than 20 mm. The median gestational age at prenatal detection was 33 weeks with a median size of 40 mm, and 18% had a complex appearance. At the time of neonatal surgery, 56% of the cysts were found to have undergone torsion, 6% were hemorrhagic, and 38% were uncomplicated. This report confirms the high risk of fetal/neonatal ovarian loss with large fetal ovarian cysts. Additional reasons for cyst aspiration (other than torsion risk reduction) are analysis of cyst fluid to confirm ovarian hormonal status and to decrease fetal abdominal dystocia at delivery with very large cysts. Arguments against the cyst aspiration are risks of bleeding, infection, and spread of malignancy. The evidence-based management plan for prenatally diagnosed ovarian cysts is not well defined and may require a multicenter RCT to establish the optimal risk:benefit outcome algorithm.

Multifetal Reduction (MFR) and Selective Termination

These ultrasound-guided techniques have been used to improve the pregnancy outcomes in higher-order multiples (triplets or more) and monochorionic twinning pathology with significant morbidity/mortality discordant birth defects.¹⁰⁴ Changes in MFR procedures have been reported for the technique, which has been available for over 20 years. A series of 2000 patients who had undergone MFR showed evolving trends with the procedure, reflecting the changes in assisted reproductive techniques and patient demographics. MFR trends have moved toward reduction of higher-order multiples to a singleton, reduction from twins to a singleton, and a strong preference for prenatal diagnosis by CVS prior to the MFR. The MFR technique uses a transabdominal injection of potassium chloride into the region of the fetal thorax under ultrasound guidance. The CVS technique is an ultrasound-guided transabdominal technique to obtain placental chorionic tissue for chromosomal and genetic analysis of each fetus prior to reduction selection. An ethical/clinical debate surrounds the use and consequences of assisted reproductive technology due to the increased incidence of multiple gestations, monochorionic twinning, and fetal anomalies. The background rate of fetal anomalies may be higher in the infertility population, but this finding is only revealed with the assisted reproductive conceived pregnancies.

Other Case Reports of Ultrasound-Guided Fetal Therapy Techniques

A 25-week fetus was identified to have a 50 × 40 mm cystic tumor originating from the sublingual area and protruding from the mouth.¹⁰⁵ The cystic tumor (60 × 60 mm) was decompressed by ultrasound-guided percutaneous needle aspiration at 26 weeks. The aspirated fluid was found to have amylase levels consistent with a ranula (salivary gland tumor). Fetal maxillary movements were observed after decompression of the cystic mass. This decompression allowed a full-term vaginal delivery of 2840-g female with an oral mass of 60 × 70 mm. Definitive surgery was completed at 25 days of life with surgical excision of the salivary gland cyst and tongue reconstruction. This decompression technique could be used in conjunction with an EXIT delivery to secure the neonatal airway/trachea, dependent on the tumor location or size due to the risk of neonatal airway obstruction.

Fetal hydrocephalus and intrauterine cerebral ventriculoscopy was considered in the early 1980s but failed to improve perinatal outcomes, and a voluntary moratorium was placed on the procedure. Newer neonatal neuoroendoscopic approaches have allowed this in utero treatment to be reconsidered. An animal (sheep) model has been developed for cerebral ventriculomegaly with results showing decreased ventriculomegaly without significant inflammatory or fibrosis of the ependyma compared to controls after shunt treatment.¹⁰⁶ More recently, the use of in utero treatment of the human fetus has been reported with an open hysterotomy technique, but without recognized improved perinatal outcomes. Four cases were reported from 1999 to 2003 using placement of a ventriculoamniotic shunt through a hysterotomy approach in the second trimester of pregnancy as treatment for isolated hydrocephalus.¹⁰⁷ The report indicated that the shunt can be inserted in a standard fashion and function well until delivery. In the 3 surviving infants, no apparent improvement in neurological function was seen as all had developmental delay consistent with their congenital hydrocephalus. Patient/fetal selection continues to be an obstacle for this proposed therapy, even with new fetal imaging and molecular genetic techniques. Fetal cerebral ventricular shunting for ventriculomegaly does not appear to improve perinatal outcomes.

Other forms of ultrasound-guided fetal therapy may include intravascular or intraperitoneal injection of genetically altered cells or stem cells to correct genetic disease, organ dysfunction, or other disease states.