In utero Treatment of Congenital High Airway Obstruction Syndrome via Fetal Laryngoscopy and EXIT Procedure Open Access

• Congenital high airway obstruction syndrome (CHAOS) is a rare and generally perinatal lethal condition, with high chance for perinatal morbidity and mortality even if the airway is secured via tracheostomy at the time of ex-utero intrapartum treatment (exit) procedure.

• Fetoscopic evaluation and attempted decompression of the obstructed airway in order to salvage the pregnancy and pulmonary development has been described, but descriptions are limited to isolated case reports and series.

• Published case reports generally describe a high rate of premature delivery following fetoscopic intervention for CHAOS, with few reports describing long-term postnatal clinical outcomes.

• Near-term delivery following fetoscopic intervention for CHAOS is feasible with subsequent long-term survival to decannulation.

• Formal EXIT procedure at the time of delivery should be considered even in the setting of a prior successful fetoscopic decompression of the fetal airway obstruction.

Congenital high airway obstruction syndrome (CHAOS) is a rare and severe birth defect that is lethal if left untreated [1, 2]. The airway obstruction may be due to several causes, including laryngeal atresia, tracheal atresia, stenosis or web, or the presence of a tracheal cyst. If left untreated, complete airway obstruction uniformly leads to large, echogenic, fluid-filled hyperinflated lungs, a flattened diaphragm, and as the disease progresses, the fetus eventually develops hydrops with high rates of stillbirth. Fetuses that survive delivery succumb to rapid respiratory failure. Even if an airway is secured at birth, the lung parenchyma displays poor oxygen exchange secondary to disrupted lung development. Traditional management strategies focused on delivering the fetus prior to the onset of severe hydrops and securing an airway via an ex-utero intrapartum treatment (EXIT) procedure. Unfortunately, this approach is still plagued with poor rates of survival. The fetuses that have survived had partial obstruction (stenosis) or complete obstruction with spontaneous perforation of the obstruction in utero [3, 4]. Herein, we report a combined fetal treatment approach in which we first performed a fetoscopic direct decompression of the trachea via perforation of the obstruction using a laser followed by delivery via EXIT to tracheostomy 11 weeks later.

A 36-year-old primigravida was referred to our service at 20 4/7 weeks’ gestation for evaluation. An ultrasound revealed normal cardiac anatomy, with dilated distal trachea (4 mm diameter), large, echogenic lungs bilaterally, flattened diaphragm, a two-vessel cord, and fetal ascites (shown in Fig. 1). Sonographic neck and chest findings demonstrated dilated trachea and bronchi distal to the fetal larynx (presumed point of obstruction) with no evidence of fluid movement across the larynx, suggesting severe obstruction. These findings were consistent with CHAOS with concern for early progression to fetal hydrops. An MRI was performed at 21 5/7 weeks’ gestation, showing laryngeal narrowing with dilation of the distal airway (shown in Fig. 2). Markedly increased lung volumes were seen with a flattened diaphragm. Left lung volume was 50 mL and right 30 mL. Observed-to-expected total lung volume ratio was 3.65. An amniocentesis was performed, resulting in a normal 46, XY karyotype and a normal male prenatal microarray. Additional targeted testing for Fraser syndrome was performed and was negative.

The patient was counseled about all management alternatives including expectant management with comfort care measures after birth, termination of the pregnancy, expectant management with an EXIT delivery with full resuscitative support, or fetoscopic laryngoscopy and possible laser treatment to open the obstructed trachea followed by an EXIT delivery with full resuscitative support. The latter procedure is minimally invasive and performed with maternal sedation and local anesthesia. This intervention can serve a dual purpose of better defining the diagnosis/prognosis and for possible treatment. Fetal laryngoscopy can be useful to assess the fetal larynx and trachea and to determine the type of obstruction. The airway obstruction may be due to laryngeal atresia, stenosis, tracheal cyst, or a membrane-like web. Cases with vocal cord fusion and segmental tracheal atresia or stenosis are typically not amenable to fetal treatment and have a grave prognosis. Cases with a thin subglottic membrane can be successfully treated via laser photocoagulation of the membrane and may have a more favorable prognosis. In addition to a frank discussion regarding the procedure-related risks and benefits of intervention with the parents, it was also imperative to acknowledge the potential for long-term respiratory, alimentary, and language development issues inherent in these medically complex patients. The parents were counseled on all management options, associated risks, and potential outcomes following intervention prior to consenting to the procedure. The patient opted to proceed with fetoscopic laryngoscopy and possible laser treatment to open the obstructed trachea.

After obtaining verbal and written consent, the patient was taken to the operating room at 24 0/7 weeks’ gestation. An amniocentesis needle was used to provide a fetal intramuscular injection of fentanyl (1 µg/kg) and rocuronium (1.5 mg/kg). Next, operative fetoscopy was performed using standard principles described previously [5‒7]. Under direct ultrasound guidance, a 3.8-mm cannula with a trocar (Richard Wolf, Vernon Hills, IL, USA) was inserted percutaneously into the amniotic cavity (e.g., direct entry method). The trocar was removed, and a 3.3-mm 30°-rigid diagnostic endoscope (Richard Wolf) was used to identify and gain access into the fetal mouth. The cannula was advanced into the fetal mouth and oropharynx, where a normal appearing fetal pharynx and glottis were identified (shown in Fig. 3).

The diagnostic scope was then exchanged for a 3.3-mm 0°-operative endoscope. While it is possible to utilize an operative endoscope for the entire procedure (anatomic mapping plus laser perforation of laryngeal obstruction), at our institution, we routinely use a rigid diagnostic endoscope (30° or 70°) to initially map the fetal anatomy and identify the target area for laser application. Then, with a 3.8 mm cannula held in position within the fetal oropharynx at the larynx, the diagnostic endoscope was exchanged for the 0° operative scope with laser fiber threaded through the working channel for subsequent laser application. To improve visualization, intermittent lavage was performed using a warmed normal saline infusion through the operating channel of the endoscope. The endoscope was advanced through the vocal cords, which appeared thickened. Just below the vocal cords, a bulging thin horizontal tissue layer was noted that obstructed the tracheal lumen (shown in Fig. 3). The 600-µm diode laser fiber was advanced to that tissue layer toward a pinpoint indentation on the surface of this tissue layer. The contact laser fiber was used to deliver short bursts of energy at 30 W. As soon as the fiber penetrated through this tissue layer, there was immediate and copious egress of secretions. Multiple punctures of the tissue were done with the laser fiber to form a small opening to the trachea. However, this opening was not large enough to allow the admittance of the endoscope. Simultaneous ultrasound showed hyperechogenic spots in the lung parenchyma periphery after the obstruction was relieved, which was attributed to microbubbles from the saline infusion. At this point of the procedure, endoscopic visualization became suboptimal due to copious egress of mucous and blood from the newly created tracheal opening. All instruments were removed. The total operative time was 71 min.

Postoperative ultrasound 24 h later showed less dilation of the trachea (3 mm diameter), but the remaining finding were relatively unchanged. Patient was discharged and followed with serial ultrasounds. A postoperative ultrasound performed at 25 weeks’ gestation showed persistent increased lung echogenicity, but the diaphragm had returned to a concave contour with a marked decrease in fetal ascites (shown in Fig. 4).

At 28 weeks’ gestation, a repeat MRI was performed, showing a stable, prominent caliber of the distal airways with decreased laryngeal narrowing and decreased eversion of the bilateral hemi-diaphragms. There was interval resolution of the abdominal ascites (shown in Fig. 5). At 33 weeks’ gestation, a repeat ultrasound showed normal lung contours and volumes with complete resolution of the fetal ascites.

At 36 weeks’ gestation, the patient was taken to the operating room in conjunction with pediatric otolaryngology (ENT) for the EXIT procedure. Combined spinal and general anesthesia was used. The patient was placed in a low lithotomy position to allow midline access for the pediatric ENT team. After delivery of the fetal head and arms with maintenance of placental support, laryngoscopy was performed by the ENT team utilizing an 8-cm Parson laryngoscope, at which time, a large obstructive tracheal cyst with a pinhole opening from the in utero procedure was observed. An attempt to dilate the opening to place a 2.5-mm diameter endotracheal tube was unsuccessful. A standard neonatal tracheostomy was then performed at the second to fourth tracheal rings. A large amount of clear yellowish fluid was found in the trachea, which was a larger caliber than expected. The total time from hysterotomy to full delivery and tracheostomy was 16 min.

The newborn was handed to neonatology. Apgar scores were 4 and 7 at 1 and 5 min, respectively. Cord gases were analyzed and demonstrated a venous pH of 7.4 with base excess of −6.4 and arterial pH of 7.3 with base excess of −5.5. Birth weight was 2,535 g (33rd percentile for gestational age). The remaining physical exam was normal.

Mechanical ventilation was maintained for 10 days, and pressor support was used for 2 days, after which he weaned to tracheostomy mist of 5 L and 21% oxygen. The total parenteral nutrition was required for 14 days. He transitioned to oral alimentation with fortified feeds at 3 weeks of life. He was discharged from the NICU on day of life #42 with tracheostomy on room air and tolerating full oral feeds.

The patient underwent interval laryngoscopic/bronchoscopic procedures with tracheostomy changes during the first 2 years of life (shown in Fig. 6), in addition to participation in physical/occupational therapy. He underwent definitive tracheolaryngoplasty and decannulation shortly after his third birthday (shown in Fig. 7) and is currently stable on room air, tolerating oral alimentation and is able to phonate. He continues to be followed by pediatric otolaryngology and participates in occupational and speech therapy.

From birth until 38 months (decannulation surgery), the patient communicated exclusively through American Sign Language; however, since decannulation, the patient has preferred spoken communication. At 40 months of life, the patient demonstrated average range receptive and expressive language based on the Developmental Assessment of Young Children-2 and Preschool Language Scales-5 by using a combination of American Sign Language and spoken words. At the time of 40 month, Bayley-4 assessment (an instrument that addresses development across 5 domains: cognitive, language, motor, social-emotional, and adaptive), the patient attained an age equivalent of 41 months and a standard score of 100 (average range). While the patient does demonstrate hoarse voice dysphonia with speech delays attributed to recent decannulation (Goldman-Fristoe Test of Articulation-r standard score of 53 falling in the delayed range), generating approximately 25% intelligible speech at this time, the patient’s overall learning efficiency and cognitive skills based on summative assessment fall in the above average range.

The patient described provided written informed consent for publication of this case report and associated images. This study protocol was approved by the local committee on human research.

Historically, management options for prenatally diagnosed CHAOS included pregnancy termination, expectant management, or EXIT at delivery, all of which generally resulted in poor outcomes unless spontaneous tracheal recanalization in utero occurred. High-resolution ultrasound imaging demonstrating findings consistent with a prenatal diagnosis of CHAOS (e.g., large, echogenic lungs, dilated distal trachea, diaphragm flattening secondary to the mass effect of large lungs, etc.) is a key in identifying potential candidates for fetal intervention. Patients with evidence of early or overt fetal hydrops represent the patient population at the highest risk for intrauterine decompensation and demise who may benefit the most from fetal intervention efforts to decompress the fetal airway. Additionally, formal fetal echocardiography and fetal MRI may be helpful in identifying additional cardiopulmonary or other anatomic malformations and in characterizing the nature of the airway obstruction (e.g., tracheal or laryngeal web vs. segmental abnormality), aiding the fetal surgery team in selecting ideal candidates for our described approach.

While EXIT followed by tracheostomy or intubation is technically feasible, mortality remains high due to the presence of hydrops with abnormal pulmonary development and postnatal complications [2, 8‒10]. Open fetal tracheoplasty for prenatal CHAOS has been reported but with long-term postnatal morbidity in the setting of a maximally invasive maternal procedure [11]. Here, we describe successful fetoscopic decompression of the fetal trachea in a pregnancy with prenatally diagnosed CHAOS, resulting in eventual reversal of the fetal hydrops and a near-term delivery by the EXIT procedure. Furthermore, long-term postnatal follow-up resulting in definitive tracheolaryngeal reconstruction and decannulation is described. While complete airway patency was not achieved prenatally, the tracheal obstruction was relieved to a sufficient extent, allowing egress of fluid such that pulmonary development was salvaged and fetal deterioration was averted.

Fetoscopic evaluation of the fetal airway with subsequent intervention has been described for a variety of indications including: (1) fetal intubation attempt prior to delivery for prenatally diagnosed airway obstruction in lieu of the more invasive outright EXIT procedure [5, 12], (2) fetal endoscopic tracheal balloon occlusion in the setting of moderate or severe congenital diaphragmatic hernia [13‒15], and (3) surgical recanalization or decompression of the fetal airway in the setting of CHAOS, tracheal cyst or bronchia atresia [16‒21].

While the aforementioned case reports have described instances such as fetal intubation to avoid EXIT procedure in the setting of an unstable fetal airway or the ability to secure a definitive fetal airway at the time of fetal laryngoscopy and intervention for prenatally diagnosed CHAOS [5, 12, 19, 21], given the nature of this patient’s airway malformation, it was not felt that additional fetoscopic intervention would have yielded any benefit to secure a definitive airway and avert an EXIT procedure. This was confirmed at the time of near-term delivery, when an oral airway was unable to be secured by ENT services, and tracheostomy was ultimately performed. Again, the primary goal of the initial fetoscopic procedure was to decompress the fetal airway in order to reverse the fetal hydrops and achieve pregnancy latency. Had fetal hydrops recurred remote from delivery, a second procedure to achieve airway decompression could have been considered.

The benefit of the fetoscopic airway evaluation, aside from being a minimally invasive approach, is that this procedure can prove to be both diagnostic and therapeutic, whereby surgical recanalization of the airway may ameliorate the risk for abnormal pulmonary development and allow for increased latency of the pregnancy with subsequent delivery near or at term. While surgical recanalization of the airway has been shown to be effective, published works are limited to case reports and series, limiting the degree to which one can estimate the chances for procedural success or intact survival [2, 6, 16, 19, 22, 23]. These published reports on successful prenatal fetoscopic intervention suggest the risk for subsequent premature delivery and need for EXIT at the time of delivery remain high, the latter of which may be dictated more so by the nature of the obstruction characterized at the time of fetoscopy. Additionally, the need for long-term ventilatory assistance and/or tracheostomy is significant, as only a few cases have described a ventilator-free postnatal course [16] or survival to decannulation [22], as we have described here in the current report.

Fetoscopic evaluation of the fetal larynx, whether or not laser or mechanical recanalization of the airway can be performed, can provide important prognostic information for the parents and neonatal care providers [20]. Additional benefits of fetoscopic evaluation of the fetal larynx and airway may include identification of other associated and potentially devastating fetal malformations such as a tracheoesophageal cleft or other tracheoesophageal anomaly [24]. Such information may also inform a parent’s decision to continue or terminate the pregnancy in question. Potential issues include the nature of the obstruction and whether this is amenable to laser perforation or mechanical dilation. Segmental abnormalities are generally not amenable laser ablation; however, the application of ultrasound-guided, fetoscopic-assisted, needle decompression of the distal airway with reversal of hydrops resulting in salvage of the pregnancy and term delivery by EXIT procedure at term has been described [25].

High-resolution ultrasound imaging, formal fetal echocardiography, and fetal MRI can be helpful in characterizing the nature of the airway obstruction and in selecting good candidates for our described approach. Imaging evaluation to rule out other congenital abnormalities in addition to diagnostic genetic analysis with chromosomal microarray to exclude genetic syndromes, in particular Fraser syndrome, which has been associated with CHAOS, is advised prior to proceeding with any diagnostic or therapeutic procedure [26‒28]. Prenatal multidisciplinary counseling by maternal fetal medicine, fetal surgery, neonatology, and experts in the management of the neonatal/pediatric airway (e.g., pediatric ENT) at a center of expertise is paramount. Technical considerations for fetoscopy include placental location for surgical access, as well as maternal size and the presence of comorbidities that may preclude the ability to perform prenatal or intrapartum intervention.

Patients without prenatal intervention should deliver by obligate EXIT procedure; however, even in the setting of successful prenatal intervention with surgical recanalization of the airway and subsequent reversal of fetal hydrops, an EXIT procedure may still be advised due to the unstable nature of the airway [23, 26, 29]. Up front risks of fetoscopy include failed tracheoplasty or the inability to access the fetal airway, as well as risk for iatrogenic premature rupture of the membranes with potential risk for subsequent premature delivery. Maternal risks associated with EXIT procedure (e.g., the need for general anesthesia, risk for placental abruption, hemorrhage, etc.) and the associated risks for fetal hypoxia have been well described [26, 27, 30].

In summary, fetoscopic airway evaluation and surgical recanalization in the setting of CHAOS is technically feasible in well-counseled, select patients, and may result in salvage of the pregnancy and reversal of fetal hydrops, with the potential for delivery at or near term. When successfully performed, this procedure optimizes the chance for pregnancy latency and long-term intact survival. Multidisciplinary counseling and management at a center of expertise with resources for postnatal management of a complex, neonatal airway remains essential for this high-risk patient population.

This study protocol was reviewed and approved by the Institutional Review Board of the University of Southern California for the Health Sciences campus, approval number HS-16-00468. Written informed consent was obtained from the patient for publication of the details of their medical case and any accompanying images.

The authors have no conflicts of interest to declare.

No external financial support was received for this work.

Rana Sabra participated in drafting and critical revision of the manuscript and interpretation and analysis of the data. Dr. Martha Monson participated in writing and critically revising the manuscript. Dr. Gheorghe Ciprian assisted in chart review, contributing example prenatal diagnostic and operative images, and writing and reviewing figure legends. Jinnen Masri assisted with chart review and data abstraction from the medical record. Dr. Ramen Chmait assisted in the conception of the study, critical revision of the manuscript, and contributing example prenatal diagnostic and operative images. All the authors reviewed the final version of the manuscript.

The data or clinical details described within this case report are not publicly available as aspects of this case report were distilled from the medical record of the research participant (pregnant individual and child). Deidentified data may be requested from the corresponding author (Ramen H. Chmait) upon reasonable request.