Fetal Surgery for Congenital Pulmonary Airway Malformation

Congenital pulmonary airway malformations (CPAMs) are lung lesions that result from disordered development of the lower respiratory tract. They were formerly known as congenital cystic adenomatoid malformations (CCAMs). CPAMs are characterized by airway cysts of varying size that are connected to the tracheobronchial tree. ^[1]  Although an elaborate postnatal staging system exists, antenatal diagnosis focuses on the size of the lesion—that is, on whether the cysts are smaller than 5 mm (microcystic) or larger than 5 mm (macrocystic). ^{[2, 3]}

CPAMs have a wide spectrum of severity and vary substantially in size and composition. ^{[4, 5]}  They are almost always unilateral and can be associated with other lung lesions, such as bronchopulmonary sequestrations, bronchogenic cysts, and congenital lobar emphysema. ^{[6, 7, 8]}  The arterial and venous blood supply of CPAMs is derived from the pulmonary circulation; however, hybrid lesions with elements of  bronchopulmonary sequestrations may have connections with the systemic circulation.

CPAM is the most common congenital lung lesion. ^{[9, 10]}  Registry data suggest that CPAMs may affect as many as 1 in 2500 live births. ^{[1, 11, 12]} This may be a slight underestimate of the true incidence of the disease, given the existence of nonregistered in-utero mortality, such as the “hidden mortality” that was seen with congenital diaphragmatic hernia. ^[13]

The underlying cause of CPAM remains unknown, but several hypotheses have been advanced. A prevailing theory is that it is caused by airway obstruction. The different presentations and lesion types are accounted for by the timing and location of obstruction. ^{[1, 14]}  Another theory is that imbalance between cell proliferation and apoptosis during airway branching morphogenesis may lead to CPAM. ^{[15, 16]} Dysregulation of the genes HOXB5 ^[17] and glial cell-derived neurotrophic factor ^[16] have also been implicated in the pathogenesis. An underlying malignancy, particularly pleuropulmomary blastoma that can be confused with macrocystic CPAM, is extremely rare among fetal lung lesions. ^{[18, 19]}

The expanded use of antenatal ultrasonography (US; see the images below) at 20 weeks' gestation has increased the detection of CPAMs and has helped characterize their natural history. ^{[6, 9, 10, 11]}  On the basis of numerous studies over the past two decades, several trends have been described.

Sonographic images of fetal congenital pulmonary airway malformation. (A) Microcystic congenital pulmonary airway malformation, with ultrasound showing a solid echogenic mass. (B) Macrocystic congenital pulmonary airway malformation, with one or more cysts >5 mm.

View Media Gallery

First, microcystic lesions can have a rapid growth phase upon diagnosis at 20 weeks’ gestation and usually peak in size at 25-26 weeks. ^[20] Subsequently, relative growth of these lesions often plateaus or, in many cases, regresses in size because of catch-up in thoracic cavity growth. ^{[21, 22, 20]} Fewer than 10% of CPAMs are undetected by US during the third trimester, but such lesions are rarely large at diagnosis. ^{[23, 24, 25]} Macrocystic lesions are less common, have more unpredictable growth patterns, and tend to persist on late gestation scans. ^[21]

Second, very large CPAM lesions that are either microcystic or macrocystic can compress other thoracic structures, such as the esophagus, mediastinum, and inferior vena cava (see the image below). This compression can result in impaired venous return, polyhydramnios, and hydrops fetalis, as defined by presence of ascites, pleural effusion, pericardial effusion, and/or skin edema. ^[4] Evidence of right-heart failure on fetal echocardiography confirms the diagnosis of hydrops. ^[26] The small percentage of CPAMs with hydrops are associated with a high mortality, and fetal intervention may be considered in these patients. ^[4]

Diagram of cystic lung mass compressing the lung and displacing the mediastinum. Illustration by Colin Fahrion, University of California-San Francisco.

View Media Gallery

The ability to predict which fetuses with CPAM will develop hydrops is critical for formulating appropriate surveillance and therapeutic strategies. Although mediastinal shift and eversion of the hemidiaphragm are markers of substantial mass effect, the most reliable predictor of hydrops risk is the CPAM volume ratio [CVR], which is the calculated volume of the lesion multiplied by 0.52 and divided by the fetal head circumference. ^{[27, 28, 29]} Seminal work in this area has shown that an initial CVR greater than 1.6 suggests hydrops development and warrants close antenatal surveillance (eg, 1-2 scans/wk). ^{[4, 27]}

Other studies have shown that an initial CVR of 1.4-2.0 is associated with a high positive predictive value for hydrops. ^{[19, 26]} Differences in the CVR threshold for hydrops are likely related to referral bias and differences in the gestational age at measurement. Regardless, early fetal care center referral of fetuses with CVRs higher than 1.0, especially if the lesion is increasing in size, is recommended; antenatal therapy may be considered if the fetus begins to show signs of hydrops fetalis. In the absence of hydrops, a CVR higher than 1.0 is associated with an increased likelihood of symptomatic disease and need for neonatal lung resection. ^{[29, 19, 30]}

The use of antenatal MRI in the diagnosis and management of CPAM is controversial and should be considered on a case-by-case basis. The additional diagnostic information is often minimal for smaller lesions. Larger CPAMs may benefit from MRI to assist with further antenatal interventions, if indicated. Moreover, there are some data to suggest that CPAMs with substantial mass effect may be associated with postnatal pulmonary hypertension and need for additional support. ^[31]

Steroid therapy

Maternal steroid therapy (12.5 mg/day IM for 2 days) is clearly indicated for microcystic CPAMs that have resulted in hydrops, especially before 26 weeks. ^[3] A low maternal-fetal risk profile is associated with this intervention, and data from several studies indicate that it is more efficacious (>90%) than fetal resection. ^{[3, 32]} Repeated steroid doses are less effective but may still be useful in selected cases. ^[33] One can only speculate on the mechanism by which steroids exert this therapeutic effect, but it is believed that they may facilitate the transition of the lesion from the canalicular to the saccular stage of development, in part by inhibiting abnormal mesenchymal proliferation within the CPAM.

Uncertainty remains about nonhydropic fetuses with large microcystic CPAMs. A randomized trial led by the University of California, San Francisco (UCSF), attempted to address the use of steroids in this context, but recruitment was hampered because of the increasing use of maternal steroids for nonhydropic microcystic CPAMs. Approximately 10% of fetal lung lesions are managed with steroids, and at some centers, steroids are routinely administered for microcystic CPAMs with CVRs as low as 1.0. ^[34] There is also evidence for reduced efficacy of steroids in macrocystic lesions, implying a different underlying developmental biology. 

Catheter-based therapies

Although steroids may be effective for reducing the size of some large macrocystic lesions, nonresponders with macrocystic disease, especially in the case of a large dominant cyst, are best treated with catheter-based therapies. ^[4]  Drainage of macrocystic CPAMs decreases the size and compressive effects of CPAMs. ^[35]  Thoracoamniotic shunts may also be indicated for hydropic fetuses with an associated large pleural effusion.

Simple cyst aspiration usually precedes thoracoamniotic shunt placement to ensure the efficacy of CPAM fluid removal. ^[36] Gestational age should be greater than 20 weeks and less than 32 weeks. The risk of chest-wall deformity is extremely high if catheter-based interventions are pursued before 20 weeks. After 32 weeks, delivery and neonatal resection are indicated rather than antenatal therapy.

Shunting procedures are performed using a 2.5 mm trocar under US guidance, with minimal disruption to the uterus and amniotic sac. Shunt choice also changes with gestational age. Before 24-25 weeks, Harrison shunts (Cook Medical, Bloomington, IN) are preferred because of their smaller size. After 25 weeks, Rocket shunts (Rocket Medical, Hingham, MA) are preferred because of the decreased likelihood of migration and the greater ease of insertion. Both shunts have a double pigtail design that helps minimize shunt dislodgment and can remain in situ until birth.

For microcystic lesions, catheter-based ablation techniques, such as radiofrequency ablation (RFA) or electrocoagulation, have been attempted with mixed success. ^[37] Unfortunately, most of these approaches tend to cause iatrogenic injury through collateral damage or edema. As a result, reported outcomes have been poor, and enthusiasm for these approaches remains low at most centers..

Fetal resection

Antenatal resection of CPAMs in the previable or borderline viable fetus was first described in the early 1980s and remains an option for very large lesions with hydrops. ^[38] However, the procedure is highly invasive and is clearly considered a second- or third-line of fetal intervention. Fortunately, fetal resection is now very uncommon, even in the largest fetal care centers in the country, owing to the efficacy of maternal steroid administrations and catheter-based interventions.

Given the high risk for complications, including maternal exsanguination and preterm labor, performance of this procedure requires a highly skilled multidisciplinary team. The fetus should be free of other anomalies, and a karyotype must be obtained to rule out a concomitant fatal aneuploidy. The mother should also be counseled of her personal risk and excluded from fetal surgical intervention if she does not meet stringent criteria.

During the procedure, deep general anesthesia is used to decrease uterine tone and preserve maternal-fetal gas exchange at the placental interface. ^[4] After a low transverse incision is made on the maternal abdomen, the position of the placenta is mapped before uterine opening with a specially designed stapler containing lactomer staples. The stapler helps create an 8- to 10-cm bloodless hysterotomy and maintains the integrity of the fragile chorion and amnion in the setting of complete uterine relaxation.

The procedure requires continuous uterine distention with isotonic fluids, as well as fetal intravenous (IV) access to enable volume expansion and cardiac support. A standard transverse anastomosis (TA) stapling device is often used once the mass is exteriorized through the thoracotomy incision. 

EXIT procedure

In rare cases, hydrops and fetal compromise can occur in or persist beyond 32 weeks despite maternal steroid therapy and catheter-based interventions. The CVR remains elevated, and significant respiratory distress is anticipated at birth. Under such circumstances, one option is the ex-utero intrapartum treatment (EXIT) procedure, which uses the placenta for gas exchange so that the fetal lungs can be bypassed and a lung resection performed before the umbilical cord is cut. ^{[39, 40]}

This approach, known as EXIT-to-resection, is based on the rationale that the native lungs will ventilate and function much better once the mass is removed. Reported outcomes following EXIT-to-resection have generally been quite favorable, though the indications for the procedure are not yet fully defined. ^[41] Extracorporeal membrane oxygenation (ECMO) may also be required, either as a bridge to resection or in the immediate postnatal period to treat residual pulmonary hypertension. ^[42]

Catheter-based therapies

Contraindications for macrocystic drainage include a predominantly solid CPAM, an abnormal fetal karyotype, severe fetal cardiac abnormalities, or lack of a window for uterine access. Other associated congenital abnormalities may also represent contraindications but must be considered on a case-by-case basis.

Fetal resection

Contraindications for fetal resection include significant maternal operative risk, an abnormal fetal karyotype, and fetal cardiac abnormalities. Other associated congenital abnormalities may also represent contraindications but must be considered on a case-by-case basis.

EXIT procedure

The EXIT procedure requires maternal laparotomy and hysterotomy. ^[39] If maternal health is significantly endangered by these interventions, EXIT should not be performed.

Maternal steroid therapy in the setting of CPAM with hydrops is assocated with an efficacy of 80-90%, as defined by hydrops resolution. Microcystic lesions treated before 26 weeks tend to have a higher response rate. ^{[3, 32]} Most, but not all, fetuses with lesions treated by means of maternal steroids are symptomatic at birth and will require neonatal lung resection.

Overall survival data for thoracoamniotic shunting for macrocystic CPAM can be derived from the aggregation of several small reports of its use. In a study by Wilson, 26 of the 41 patients treated survived (63%). ^[35] A large single-center report by Peranteau et al described thoracoamniotic shunt placement in 38 patients with macrocystic lung lesions that were either causing hydrops or believed to be at high risk for causing lung hypoplasia. ^[43] Hydrops was present in 69% of the population. Overall survival in this group was 73%.

Several investigators have demonstrated a survival rate of approximately 50% for hydropic fetuses with microcystic CPAM after surgery. ^[4] Hydropic fetuses treated with steroids, however, have survival rates near 85%. ^[3] The superiority of steroid therapy, combined with its complete abrogation of serious maternal risk, has made it the standard of care for hydrops caused by microcystic CPAM, especially before 26 weeks. If hydrops persists or emerges past 32 weeks, EXIT and neonatal resection remain options. ^{[42, 40]}