Unlike obstruction of the urinary tract at other levels, bladder outlet obstruction has the potential to affect the development of the whole urinary tract as well as the lungs. In males, the most common cause of bladder outlet obstruction is posterior urethral valves (PUVs) (Atwell, 1983). In contrast, in females the most common cause of bladder outlet obstruction is urethral atresia. As in many other congenital anomalies, there is a broad range of severity, with bladder outlet obstruction ranging from completely asymptomatic to the infant who presents at delivery with respiratory insufficiency due to pulmonary hypoplasia caused by long-standing oligohydramnios and renal failure from renal dysplasia.

PUVs are thought to be embryologically derived from mullerian duct remnants or remnants of the cloacal membrane dating back to between the 7th and 11th weeks of gestation (Dewan et al., 1992, 1995). Young et al. (1919) described a classification scheme for PUVs. In type I, the valves are sail-like leaflets that arise from the crista-urethralis distal to the verumontanum. These valves may cover only the lower half of the urethra or may encircle the urethra and cause complete obstruction. Type II valves are nonobstructing folds of the superficial muscle and mucosa that extend from the verumontanum to the bladder neck (Cuckow, 1998). Type III valves usually cause a diaphragm-like obstruction at the level of the verumontanum, but can be seen at the level of the anterior urethra distal to the external urethral sphincter (Hendren, 1971).

The characteristic prenatal presentation of bladder outlet obstruction is that of a dilated bladder and bilateral hydroureteronephrosis. The severity of prenatal bladder outlet obstruction ranges from mild to severe as is seen postnatally. At the better end of the spectrum, the fetus may have obstructive uropathy with preservation of amniotic fluid volume, minimal changes in the bladder and ureters, and no dysplastic changes in the kidneys. At the severe end of the spectrum are fetuses with profound oligohydramnios, distended bladder, and ureters with cystic dysplastic changes in the kidneys. The longstanding oligohydramnios results in pulmonary hypoplasia, which leads to severe respiratory insufficiency at birth and is a leading cause of death during the neonatal period. Fetuses with obstructive uropathy can also have other associated nongenitourinary anomalies, chromosomal abnormalities, and deformations related to oligohydramnios, emphasizing the importance of careful prenatal evaluation. In cases of isolated bladder outlet obstruction due to PUVs, vesicoamniotic shunting may be lifesaving.

The figures reported for the incidence of bladder outlet obstruction vary widely, reflecting the spectrum of severity seen in obstructive uropathy. In PUVs, the most common cause, the range in severity extends from fetuses presenting with oligohydramnios, renal dysplasia, and pulmonary hypoplasia to asymptomatic octogenarians noted to have PUVs at the time of cytoscopy for benign prostatic hypertrophy. The incidence of PUVs has been reported to be between 1 in 5000 and 1 in 25,000 births (Atwell, 1983). PUVs account for 10% of all urologic anomalies detected by prenatal ultrasound, which occur as frequently as 1 in 800 livebirths (Thomas and Gordon, 1989;Hutton et al., 1994). Prenatal ultrasound picks up 50% of new cases of PUVs, suggesting an incidence of 1 in 4000 livebirths (Cuckow, 1998). Recently, there has been increased awareness of PUVs, and Richmond and Atkins reported an increase in apparent prevalence from 1.9 per 10,000 births to 2.4 per 10,000 births (Richmond and Atkins, 2005). This estimate does not take into account cases of intrauterine fetal death, loss to pregnancy termination, stillbirths, or the asymptomatic cases that present later in life with voiding difficulty (Hendren, 1971; Pieretti, 1993).

Urinary tract obstruction at the level of the bladder outlet is usually due to PUVs in male fetuses and urethral atresia in females. The cardinal features of bladder outlet obstruction include marked and persistent dilation of the urinary bladder with a thickened, often trabeculated, bladder wall (Glick et al., 1984; Mahony et al., 1985). The posterior urethra is dilated in urethral obstruction due to PUVs. This dilated proximal urethra resembles a keyhole, extending from the bladder toward the fetal perineum (Figure 82-1). The dilated bladder can become quite large, filling both fetal pelvis and abdomen. The mural thickness of the normal bladder is quite thin and bladder wall thickness greater than 2 mm is pathologic. In cases of severe bladder outlet obstruction, the nondilated bladder (either following bladder tap or spontaneous voiding) can be as thick as 10 to 15 mm (Mahony, 1994). Often, severe bladder outlet obstruction will result in ureterectasis and caliectasis due to vesicoureteral reflux induced by high intravesical pressure. However, the absence of these features does not preclude the diagnosis of bladder outlet obstruction because only 40% of fetuses will demonstrate ureterectasis and pyelectasis in bladder outlet obstruction (Mahony, 1994). In high-grade bladder outlet obstruction, the urinary bladder may spontaneously decompress through rupture of the urinary tract, resulting in either fetal urinary ascites or perinephric urinoma (Callen et al., 1983; Mahony et al., 1984). The end-stage effects of early gestation high-grade bladder outlet obstruction is often bilateral renal dysplasia. This may be sonographically evident from markedly increased renal parenchymal echogenicity and, most specifically, from the presence of subcortical cysts (Mahony et al., 1984; Crombleholme et al., 1991). Paradoxically, the lack of caliectasis in the otherwise obstructed urinary tract may suggest the presence of renal dysplasia and reflects the lack of urine production by severely damaged kidneys. Oligohydramnios is indicative of high-grade obstruction and, if long-standing, may result in deformations including Potter facies and clubfeet.

Fetal MRI has been found to be a useful adjunctive imaging modality in the evaluation of the fetus with obstructive uropathy (Caire et al., 2003). MRI may be more sensitive in defining subcortical cysts, suggesting renal dysplasia. In addition, fetal MRI may provide anatomic detail on pelvic structures not available with ultrasound, particularly in the setting of bladder outlet obstruction presenting prior to 16 weeks of gestation. In these early cases, persistent cloaca may be more clearly diagnosed by MRI than by ultrasound alone.

The functional capacity of the fetal kidney affected by obstructive uropathy depends on the extent and severity of renal dysplasia caused by the obstruction. The dysplastic fetal kidney is characterized by the presence of disorganized metanephric structures surrounded by fibrous tissue which, in addition, may have cortical cysts (Rubenstein et al., 1961; Beck, 1971; Potter, 1972;Risdon, 1975; Bernstein, 1976). More than 90% of dysplastic kidneys with cortical cysts are associated with obstruction during nephrogenesis (Rubenstein et al., 1961; Bernstein, 1976). Detection of cortical cysts sonographically implies the presence of severe renal dysplasia and irreversible renal damage, excluding the patient from intervention. A normal kidney will display an echotexture similar to that of the liver, with an internal architecture showing a differentiation between cortex and medulla. The medulla containing tubules and fluid will appear darker. A dysplastic kidney, however, will show no internal architecture and may display an increased echogenicity due to a disruption in the normal renal histology. Renal dysplasia may be associated with cystic formation with the parenchyma (Harrison et al., 1982a; Risdon, 1975). Features of multicystic dysplasia of the kidney (MCDK) include the presence of multiple noncommunicating cysts of variable sizes, interfaces between the cysts (presence of echogenic areas within the renal parenchyma), nonmedial location of the largest cyst, and absence of an organized parenchyma (see Chapter 78) (Fong et al., 1986; Kleiner et al., 1986). Multicystic dysplasia of the kidney is most often seen unilaterally and is associated with a high incidence of contralateral urologic anomalies that will warrant postnatal evaluation by a pediatric urologist (Thomas, 1990). Cystic dysplasia should not be confused with severe hydronephrosis.

Mahony et al. (1984) studied the kidneys of 49 fetuses with obstructive uropathy and found that the presence of cortical cysts had a positive predictive value of 100% for the presence of renal dysplasia. Sonography was also 100% specific as no fetus without dysplasia had detectable cortical cysts. The presence of cortical cysts reliably predicts the presence of renal dysplasia and irreversible renal damage, but the absence of cortical cysts cannot ensure the absence of renal dysplasia. Of the kidneys without cortical cysts in the study by Mahony et al., only 44% were free of dysplastic changes. Renal dysplasia may be present without cysts, or the diameter of the cysts may be below the resolution of ultrasound. Technical factors may also limit the ability of the sonographer to adequately image the fetal kidneys, including shadowing from the adjacent spine and oligohydramnios, which often accompanies obstructive uropathy.

In the dysplastic kidney, there is abundant fibrous tissue that may increase the echogenicity of the renal parenchyma. It has been suggested that increased echogenicity at the renal cortex may be a sign of renal dysplasia. However, when this sonographic sign was evaluated, it was shown to be less specific and to have a lower positive predictive value than the presence of cortical cysts (Mahony et al., 1985). The evaluation of renal echogenicity is also a more subjective assessment with inherent interobserver variability, which further limits its utility.

Ultrasonographic examination of the fetal kidneys may provide prognostic information if cortical cysts and increased echogenicity are detected, but is less specific in their absence (Mahony et al., 1985; Crombleholme et al., 1991). Similarly, the volume of amniotic fluid is not a useful prognostic indicator except at the extremes (Harrison et al., 1981a, 1982a, 1987; Glick et al., 1985). Zaccara et al. found that amniotic fluid index remains a reliable predictor of renal function in cases of obstructive uropathy (Zaccara et al., 2005). Preserved amniotic fluid almost always predicts normal renal function at long-term evaluation. Fetuses with bilateral hydronephrosis and normal amniotic fluid may not require intervention. Similarly, fetuses with bilateral hydronephrosis, severe oligohydramnios, and severely dysplastic renal cortex as seen on ultrasound are unlikely to benefit from in utero therapy. It is for the fetuses between these two extremes that prognostic criteria are most important. Assessment of residual fetal renal function by indirect methods such as ultrasound determination of bladder filling and emptying or furosemide stimulation of urine production have not proven reliable (Campbell et al., 1973; Wladimiroff, 1975; Chamberlain et al., 1985). A more sensitive means of assessing fetal renal function is essential for the appropriate selection of fetuses with obstructive uropathy for treatment.

An evaluation of the fetal urinary tract should include an assessment of the overall growth and development of the fetus, amniotic fluid index, gender, renal parenchymal appearance, extent of dilatation of the collecting system, unilateral or bilateral involvement, and bladder size, thickness, and emptying (Cendron et al., 1994). Because of the increased incidence of associated malformations, the fetus should be scanned for extrarenal anomalies. Serial sonographic evaluation is also essential to evaluate for progression of these sonographic features.

The sex of the fetus will help narrow the differential diagnosis. Sonographic demonstration of male external genitalia in the setting of bladder outlet obstruction strongly suggests the diagnosis of PUVs. In a female fetus, one should consider the diagnosis of urethral atresia, persistent cloaca, caudal regression anomaly, or megacystis-microcolon–intestinal hypoperistalsis syndrome (Table 82-1). An unusual cause of bladder outlet obstruction that can occur in either sex is prolapse of an ectopic ureterocele from a duplex collecting system. In distinguishing urethral atresia in a female fetus, the oligohydramnios is profound and long standing. Particularly when diagnosed prior to 16 weeks of gestation, a dilated “bladder” on ultrasound should raise the possibility of persistent cloaca. The presence of a persistent cloaca may be suspected from the anatomy of the “bladder” and the presence of debris in the cloaca from the intestinal communication. Intraluminal calcifications within bowel loops suggest a communication between intestinal and genitourinary tracts. Caudal regression should be evident from images of the fetal spine. In the megacystis-microcolon–intestinal hypoperistalsis syndrome, the bladder is very dilated but with a thin wall even after decompression, as opposed to the pressure-induced hypertrophy observed in bladder outlet obstruction from PUVs. This is a rare, usually lethal, anomaly in which 80% of affected fetuses are females (Winter and Knowles, 1986). Other features that are clues to this diagnosis include small-bowel dilation, a microcolon with malrotation, and polyhydramnios in the third trimester (Stramm et al., 1991). This is inherited as an autosomal recessive condition and therefore has implications for future pregnancies.

A thorough understanding of the pathophysiology of fetal obstructive uropathy and its sequelae in the developing fetus is essential in formulating a rational approach to clinical management. It is important to understand why obstructive uropathy results in pulmonary hypoplasia, renal dysplasia, and associated malformations and whether or not these changes can be reversed by decompression in utero.

Obstruction of the urethra during the latter half of gestation results in dilatation and hypertrophy of the bladder and megaureter and bilateral hydronephrosis. But obstruction that starts this late in gestation does not usually produce the dysplastic changes in the renal parenchyma seen in human fetuses. Some have argued that dysplasia is caused by an abnormal interaction between the ureteric bud and the metanephric mesenchyme and is only incidentally associated with obstruction. Experiments in a fetal chick in which ureteric buds were denuded of metanephric mesenchyme formed primitive ducts, suggesting that in the chick, dysplasia may be due to nonobstructive causes. These studies have questionable relevance to the pathogenesis in higher animals and humans.

The proponents of obstruction-induced dysplasia argue that after 8 to 10 weeks of gestation, when the kidney begins to make urine, obstruction results in back pressure into the developing nephrons. Potter demonstrated that the collecting tubules of the nephrogenic kidney are straight and short and may be more susceptible to injury from back pressure than the mature nephron. Chevalier (1993) confirmed that early in development, renal injury from ureteral obstruction occurs more frequently than in the adult kidney. It follows that in order to preserve renal function in the face of an obstruction, early recognition and prompt decompression may be necessary. Early obstruction will cause the development of dysplastic changes in the kidney, but further impairment in renal function may occur not so much as a result of nephron loss from a pressure-mediated mechanism, but by a vasoconstrictive effect mediated by an overactivation of the renin–angiotensin system (Chevalier, 1993).

In humans, oligohydramnios, because of obstructive uropathy, renal agenesis, or prolonged amniotic fluid leak, results in severe pulmonary hypoplasia (Kemper and Mueller-Wiefel, 2007). The lungs of infants affected by obstructive uropathy show decreased airway branches from segmental bronchi, reflecting compromised development during the first half of gestation (pseudoglandular stage, 5 to 16 weeks). Although the lung made hypoplastic by compression during the pseudoglandular stage would not be expected to develop new airway branches, it would retain the capacity to make new alveoli and intra-acinar arteries. To evaluate the effect of obstructive uropathy on pulmonary development in an early gestation model, Adzick et al. (1987) performed bilateral ureteral ligation at 60 days of gestation, during the pseudoglandular stage of lung development (60–80 days of gestation), on lambs. Morphometric analysis of the lungs at birth revealed significant reduction of lung volume, radial alveolar count, and mean linear intercept (an indicator of airspace size) in the lambs with bilateral ureteral ligation as compared with the controls (Adzick et al., 1987). In addition, muscularization of the intra-acinar arteries was greater in lambs with bilateral ureteral ligation, indicating increased peripheral pulmonary arteriolar muscularization. These morphometric findings were the same as those in human infants who died from oligohydramnios-induced pulmonary hypoplasia from obstructive uropathy (Potter, 1972; Hislop et al., 1979). These experimental models replicated both the renal dysplastic changes and the pulmonary hypoplasia that are observed in human fetuses who die from severe obstructive uropathy and oligohydramnios.

The mechanism by which oligohydramnios causes pulmonary hypoplasia is uncertain. Several possible contributing factors include the small uterine cavity, which causes thoracic compression and limited intrathoracic space. Fetal breathing movements may be limited by uterine compression of the fetal chest and abdomen. Fetal respiration is thought to be an important factor in lung growth (Wigglesworth and Desai, 1982). Ablation of fetal breathing by cord transection, curare-induced paralysis, or damping fetal breathing movements by thoracoplasty—all result in pulmonary hypoplasia (Liggins et al., 1979; Wigglesworth and Desai, 1979; Moessinger, 1983). Oligohydramnios may also increase the loss of lung fluid, with decreased lung fluid volume within the airway (Lanman et al., 1971). Chronic fetal tracheal drainage of lung fluid results in pulmonary hypoplasia. The amniotic fluid may also act as a hydraulic stent. Increases in amniotic pressure are transmitted to the fluid in the fetal airway and prevent increases in transthoracic pressure, which might restrict lung growth from chest compression. Another possibility is the loss of a growth factor produced by the kidneys (Lanman et al., 1971).

These studies form the basis for in utero intervention to decompress the urinary tract and restore amniotic fluid dynamics to prevent neonatal death due to pulmonary hypoplasia and renal dysplasia.

A fetus with suspected bladder outlet obstruction should undergo a detailed sonographic survey to detect nongenitourinary anomalies. In particular, sonographic features of trisomies 13, 18, and 21 should be ruled out because chromosomal abnormalities can be seen in bladder outlet obstruction in as many as 12% of cases (Crombleholme et al., 1991; Cusick et al., 1995). Deformations due to oligohydramnios such as clubfoot and Potter facies should be ruled out. A sonographic evaluation of the urinary tract should include the sex of the fetus, amniotic fluid volume, presence or absence of ascites, and a detailed examination of the urinary tract itself. Not only the presence of the keyhole sign, but the size of the bladder and evidence of bladder wall hypertrophy should be noted. The presence or absence of an ureterocele should be sought, and the bladder should be observed during voiding for evidence of vesicoureteral reflux. The degree of ureteral dilation, hydronephrosis, and/or caliectasis may increase with voiding in the presence of vesicoureteral reflux. All fetuses with bladder outlet obstruction should have a genetic consultation and should consider amniocentesis. Echocardiography should be performed to rule out structural heart disease. In less severe cases of bladder outlet obstruction, in which amniotic fluid volume is preserved, there should be no adverse effects on pulmonary development and the site, timing, and mode of delivery should be unaffected. In severe cases in which oligohydramnios has occurred, either a decision should be made not to aggressively resuscitate the infant, or the baby should be delivered in a tertiary care setting. In the case of severe long-standing bladder outlet obstruction associated with oligohydramnios and renal dysplasia, there should be no intervention for nonreassuring fetal testing or attempts at resuscitation because these infants have severe pulmonary hypoplasia and renal dysplasia that are incompatible with life. Prenatal consultation with pediatric specialists in urology, nephrology, and neonatology may be helpful in counseling parents about the options for treatment and long-term outcome. The approach to the fetus with bladder outlet obstruction is outlined in the algorithm in Figure 82-2.

It is important to stress that the appearance of the fetus with bladder outlet obstruction may evolve during gestation. The fetus with complete or high-grade bladder outlet obstruction associated with oligohydramnios may spontaneously evolve with development of less-severe outlet obstruction and restoration of amniotic fluid volume. Conversely, the fetus with high-grade partial bladder outlet obstruction but preserved amniotic fluid volume may progress during pregnancy, with the kidneys becoming more echogenic and demonstrating subcortical cysts consistent with the development of dysplasia. Serial sonographic surveillance should be a part of a prenatal management plan of all fetuses with bladder outlet obstruction.