Fetal hydrothorax (FHT), either unilateral or bilateral, is a pleural effusion that may be primary, due to chylous leak, or secondary, in which the effusions are part of a generalized fluid retention associated with immune or nonimmune hydrops (Holzgreve et al., 1985; Longaker et al., 1989; Laberge et al., 1991). The management of pleural effusion in the fetus is complicated by the difficulty in distinguishing primary from secondary FHT. Secondary FHT is far more common in the fetus than in the neonate (Holzgreve et al., 1985; Hagay et al., 1993). Secondary FHT may be due to a wide variety of maternal and fetal disorders, including chromosomal anomalies, cardiovascular, hematologic, gastrointestinal, pulmonary, metabolic, infectious, and neoplastic, and malformations of the placenta and umbilical cord (Chernick and Reed, 1970; Petres et al., 1982; Gardner et al., 1983; Grena-Ansotegui et al., 1984; Van Gerde et al., 1984; Foote and Vickers, 1986; Nicolaides and Azar, 1990). While the underlying cause of a secondary FHT may be evident from detailed sonographic examination and karyotype analysis, in many instances the cause of the effusion remains obscure even after a postmortem examination (Keeling et al., 1983; Nicolaides et al., 1985).

Cases of secondary FHT may occur as frequently as 1 in 1500 livebirths (Hutchison et al., 1982; Im et al., 1984; Castillo et al., 1986). The true incidence of primary FHT is uncertain. In a review of cases from five obstetrical hospitals in Montreal between 1980 and 1987, Longaker et al. (1989) estimated that the incidence of primary FHT is 1 case per 12,000 livebirths. The actual incidence of primary FHT may be even higher if one considers that in many cases the condition may remain undiagnosed, it may resolve spontaneously, the fetus may be aborted, or death may occur soon after birth in outlying hospitals before transfer to a tertiary care center (Longaker et al., 1989; Laberge et al., 1991). In a review of reported cases of FHT, Weber and Philipson (1992) found that there was a 2:1 male to female ratio, which is similar to the ratio in neonates with chylothorax (Chernick and Reed, 1970; Broadman, 1975; Weber and Philipson, 1992).

The first sonographic diagnosis of FHT was made by Carroll in 1977, followed shortly thereafter by Defoort and Thiery (1978). Although the earliest gestation age at which prenatal diagnosis has been made sonographically is 17 weeks, the majority of cases are not diagnosed until after 30 weeks (Laberge et al., 1991). Reports of prenatally diagnosed FHT have continued to appear, and hundreds of cases have been described in the literature (Defoort and Thiery, 1978; Peleg et al., 1985; Wilson et al., 1985; Benacerraf et al., 1986; Calisti et al., 1986; Roberts et al., 1986; Weiner et al., 1986; Castillo et al., 1987; Murayoma et al., 1987; Reece et al., 1987; Adams et al., 1988; Blott et al., 1988; Bruno et al., 1988; Yaghoobian and Comrie, 1988; Longaker et al., 1989; Landy et al., 1990; Lien et al., 1990; Eddleman et al., 1991; Hernanz-Schulman et al., 1991; Parker and James, 1991).

In FHT, sonography demonstrates an anechoic space located peripherally around the compressed lungs (Figure 38-1). If the effusion is sufficiently large, there may be some degree of tension noted, with shift of the mediastinum away from the FHT and flattening or eversion of the ipsilateral diaphragm. The heart may be shifted into the contralateral hemithorax and appear smaller than normal. The presence of septations or solid components within the intrathoracic fluid collection suggests alternative diagnoses (Sydorak et al., 2002; Tsao et al., 2003). FHT has been reported in association with congenital diaphragmatic hernia, congenital cystic adenomatoid malformation of the lung, and bronchopulmonary sequestration, but it can usually be differentiated from simple hydrothorax by its more echogenic appearance (Smith, 1982; Longaker et al., 1989; Hernanz-Schulman et al., 1991; Laberge et al., 1991). Blott et al. (1988) reported FHT in association with a right-sided congenital diaphragmatic hernia and ascites. The hydrothorax resulted from a fluid-filled peritoneal sac in the right chest. The association with FHT has been reported in up to 20% of cases of congenital diaphragmatic hernia and should be considered in the differential diagnosis of FHT (Liew, 1974; Sydorak et al., 2002; Khalil et al., 2005), although the effusion in these cases is usually small.

FHT may be the first sign of nonimmune hydrops, and a careful sonographic inspection should be made to detect subtle signs of this. Congenital heart disease is observed in up to 5% of cases of prenatally diagnosed FHT (Hagay et al., 1993). Large effusions with shift of the mediastinum and compression of the heart may limit delineation of cardiac anatomy on fetal echocardiography (Rodeck et al., 1988). Evacuation of the pleural cavity by fetal thoracentesis will shift the heart to the midline, which may facilitate adequate imaging of the heart (Bigras et al., 2003).

Polyhydramnios is associated with FHT in 60% to 70% of cases (Defoort and Thiery, 1978; Peleg et al., 1985; Wilson et al., 1985; Calisti et al., 1986; Roberts et al., 1986; Weiner et al., 1986; Castillo et al., 1987; Murayoma et al., 1987; Reece et al., 1987; Adams et al., 1988; Blott et al., 1988; Bruno et al., 1988; Yaghoobian and Comrie, 1988; Lien et al., 1990; Longaker et al., 1989; Landy et al., 1990; Nicolaides and Azar, 1990; Eddleman et al., 1991; Parker and James, 1991). This often prompts referral for prenatal sonography because of discordant size and dates. The cause of polyhydramnios in FHT is not known for certain, but it has been suggested that large FHT associated with mediastinal shift may interfere with fetal swallowing (Hagay et al., 1993). This view is supported by one study that demonstrated a lack of contrast agent in the gastrointestinal tract after intra-amniotic instillation (Urograffin) (Murayoma et al., 1987). Alternatively, Petres et al. (1982) suggested that the cause of polyhydramnios in FHT may be an alteration in the production of amniotic fluid by the compressed lungs.

A thorough search should be made for associated extrathoracic malformations, such as cystic hygroma or other features suggesting Turner syndrome (Smith, 1982; Foote and Vickers, 1986). Down syndrome has been associated with congenital pleural effusion (Puntis et al., 1987). Both Turner and Noonan syndromes include malformations of lymphatic vessels, but only rarely are these syndromes associated with congenital chylothorax (Van Gerde et al., 1984).

The most important factor in the differential diagnosis of fetal pleural effusions is distinguishing primary from secondary FHT. A detailed sonographic examination will reveal a major congenital anomaly in up to 40% of cases of secondary FHT (Hutchison et al., 1982). In most instances, however, a fetal thoracentesis is needed to distinguish primary from secondary FHT (Broadman, 1975). A pleural-fluid differential cell count that consists of more than 80% lymphocytes is considered pathognomic for chylothorax, which is a characteristic finding in neonates (Puntis et al., 1987; Longaker et al., 1989; Pijpess et al., 1989). Eddleman et al. (1991) have questioned the accuracy of a lymphocyte count as an indicator of chylothorax in the presence of a viral infection. However, the serositis that results from viral infection is usually not limited to the pleural cavities, and pericardial effusion and ascites are also often observed. A predominance of atypical lymphocytes on differential cell count may also suggest a viral infection.

Chylothorax has been described in association with diffuse congenital lymphangiomatosis (Berberich et al., 1975; Roth, 1984; Smelzer et al., 1986), diffuse hemangiomata, chylopericardium (Bhatti et al., 1985), congenital lymphedema (Pearl and Wilson, 1981; Kitohara, 1985; Lev-Sagie et al., 2003), pulmonary lymphangiectasia (Jauppela et al., 1983; Hunter and Becroft, 1984; Kerr-Wilson et al., 1985), bronchopulmonary sequestration (Kristofferson and Ipsen, 1984; Hook et al., 1987), congenital diaphragmatic hernia (Liew, 1974; Sydorak et al., 2002; Khalil et al., 2005) atresia of thoracic duct, generalized pleural oozing, and other malformations (Manning and O’Brien, 1983; Lazarus and McCurdy, 1984). In many cases no underlying cause for the chylothorax is found.

The natural history of FHT is significantly different from chylothorax in the newborn and carries a much poorer prognosis. The mortality rate for chylothorax in the newborn is at most 15%, but the mortality rate for prenatally diagnosed FHT is 53% (Broadman, 1975; Longaker et al., 1989; Klam et al., 2005).

Several features of primary FHT are associated with a more favorable outcome. Unilateral FHT, without evidence of tension, such as mediastinal shift or diaphragmatic eversion, is associated with 100% survival (Longaker et al., 1989; Laberge et al., 1991). This contrasts with a survival rate of only 52% in bilateral FHT (Longaker et al., 1989). Survival is also observed in all cases of FHT that spontaneously resolve (Longaker et al., 1989; Laberge et al., 1991). Spontaneous resolution of FHT has occurred in approximately 5% to 22% of cases (Carroll, 1977; Jaffe et al., 1986; Blott et al., 1988; Yaghoobian and Comrie, 1988; Longaker et al., 1989; Landy et al., 1990; Lien et al., 1990; Nicolaides and Azar, 1990; Eddleman et al., 1991; Klam et al., 2005). Because of the possibility of spontaneous regression, a period of observation is warranted in all cases of FHT.