A 26-year-old pregnant woman (gravida 1, para 0) was referred to our institution for management of suspected fetal CHAOS after a routine detailed obstetric sonogram at 22 weeks' gestation showed fetal ascites and enlarged echogenic lungs (Fig. 1A). The four-vessel view of the heart was normal. A repeat sonogram at 24 weeks' gestation confirmed the findings of pulmonary enlargement with fetal hydrops and ascites, but dilated airways were not seen and the central airway was considered normal.

In view of these atypical features, MRI was performed at 24 weeks' gestation, confirming the presence of ascites and pulmonary overexpansion. In addition, MRI showed mildly dilated intrapulmonary airways with apparent narrowing of the central airway between the thoracic inlet and the carina (Figs. 1B and 1C). The central airway appeared otherwise patent without intrinsic occlusion (Fig. 1D). The possibility of extrinsic vascular compression was raised and was subsequently confirmed on fetal echocardiography, which showed a double aortic arch with a dominant right arch, a smaller left arch, and left ductal arch forming a ring surrounding the trachea (Figs. 1E and 1F). On pulsed Doppler interrogation, left and right ventricular inflow patterns and flow patterns in the inferior vena cava (no increase in a reversal), ductus venosus (no a wave reversal), and umbilical vein (no pulsations) appeared normal. The diagnosis was further confirmed by the presence of mirror image bifurcated brachiocephalic arteries in the same plane, cephalad to the aortic arch (Fig. 1G).

A female infant was delivered vaginally at 30 weeks' gestation after spontaneous preterm labor. The infant required immediate intubation for respiratory insufficiency. Extubation was not possible until the nondominant arch was surgically divided at 6 weeks of age. Subsequent postoperative recovery was uneventful.

Prenatal central airway obstruction can lead to fetal demise or stillbirth and can be due to intrinsic airway defects (collectively termed “CHAOS”) or extrinsic compression by neck masses or vascular rings. The recognition of CHAOS on prenatal sonography may allow life-saving airway management by the ex utero intrapartum treatment or by in utero tracheotomy [6, 7]. The impressive imaging findings that develop due to the airway obstruction in CHAOS (pulmonary overexpansion, distal airway dilatation, and fetal hydrops or ascites) have not been previously reported with extrinsic compression by tumors or vascular rings, although such congenital abnormalities are a recognized cause of postnatal airway obstruction. Prenatal sonography or MRI may depict these conditions, but polyhydramnios (presumably due to impaired fetal swallowing) is the only reported finding suggestive of prenatal airway or esophageal obstruction.

In this case, the fetus had dramatic sonographic findings that mimicked those of CHAOS and that were presumably causing airway obstruction. These findings could have prompted an inappropriate intervention such as an in utero tracheotomy. Fetal MRI was helpful in precisely localizing the level of airway narrowing and raising the possibility of extrinsic vascular compression. Fetal echocardiography revealed the specific cause of extrinsic compression—the presence of a double aortic arch.

This case highlights the need to critically review imaging findings in apparent CHAOS cases on prenatal sonography. Atypical features should lead to further imaging with fetal MRI or fetal echocardiography because these techniques may help identify extrinsic vascular anomalies, such as the double aortic arch seen in this case. Recognition that double aortic arch may mimic CHAOS on prenatal sonography may prevent misdiagnosis and facilitate appropriate treatment.

The present case is of interest for a second reason. It has been proposed that the cause of hydrops in CHAOS is secondary to cardiac compression with impedance of systemic venous (including umbilical venous) return. The absence of Doppler flow changes consistent with compression and increased central venous pressure—including dominant e wave in ventricular filling, increased a wave reversal in the inferior vena cava, a wave reversal in the ductus venosus, and umbilical venous pulsation [8, 9]—suggested another mechanism for the hydrops. Perhaps with significant lung expansion and inversion of the diaphragms, there may be impedance of lymphatic return through the thoracic duct resulting in abdominal lymphatic congestion and the evolution of ascites, which was the primary extrathoracic finding in our case.

Address correspondence to F. V. Coakley.

This is a Web exclusive article.