Fetal Surgery for Urinary Tract Obstruction

Urinary tract obstructions are a common antenatal diagnosis, and outcomes can vary from clinical insignificance to intrauterine fetal demise (IUFD). Many of these patients survive to birth but develop end-stage renal failure, requiring renal replacement therapy and ultimately kidney transplantation.

A role for fetal intervention was originally proposed in the 1980s, when Harrison et al demonstrated that urinary diversion following obstruction could inhibit the progression of renal parenchymal damage. ^[1]  However, determining which patients may benefit from urinary diversion has proved challenging and requires extensive antenatal evaluation. Currently, fetal intervention strategies for urinary tract obstruction include vesicoamniotic shunts, which can act as a temporizing measure and provide a survival advantage in a select cohort, and fetal cystoscopy, which is under investigation by multiple groups. ^{[2, 3, 4, 5]}

This article reviews the diagnosis, evaluation, and treatment strategies for fetal urinary tract obstruction.

Clinically significant urinary tract obstructions occur with a frequency of 1 in 500 live births and are associated with high morbidity and mortality. ^[6]  With antenatal ultrasonography (US), urinary tract anomalies can be detected as early as 12-14 weeks’ gestation. These anomalies can be stratified into upper urinary tract obstruction (UUTO) and lower urinary tract obstruction (LUTO).

Common causes of UUTO include ureteropelvic junction (UPJ) obstruction, ureterovesical junction obstruction, collecting system duplications, multicystic dysplastic kidney, ureterocele/ectopic ureter, and pelvic tumors. Because UUTO is not currently amenable to fetal intervention, the primary focus of this article is LUTO.

Posterior urethral valves (PUVs) account for 9% of all urinary tract obstructions and more than 50% of LUTOs. ^[7] Other causes of LUTO include urethral atresia (the second most common cause), anterior urethral valves, meatal stenosis, epispadias, and hypospadias. ^[8] LUTOs are far more common in males, and when identified in females, they should raise suspicion for cloacal malformations.

LUTO is often diagnosed with antenatal US during the first or second trimester. Recommendations from the European Reference Network for Rare Kidney Diseases (ERKNet) have proposed the anteroposterior diameter (APD) of the renal pelvis as the most reliable parameter for diagnosing suspected LUTO. ^[9] Other frequent findings are megacystis, bilateral hydronephrosis, pyelocaliectasis, and oligohydramnios or anhydramnios. The hallmark finding of PUV on US is the "keyhole sign," which results from bladder and urethral distention proximal to the valve (see the image below). The presence of a normal volume of amniotic fluid suggests a partial obstruction.

Classic keyhole sign seen in patients with posterior urethral valves.

View Media Gallery

Urinary obstructions can lead to cystic renal dysplasia and impairment of the glomerular and tubular apparatus. In severe cases, oligohydramnios or anhydramnios may ensue, predisposing the fetus to limb abnormalities and pulmonary hypoplasia. Oligohydramnios/anhydramnios associated with urethral obstruction carries a grim prognosis, and when it is identified in the early to middle stages of gestation, mortality approaches 95%. ^[10] Prognostic indicators such as bladder aspiration, evaluation of refill time, US findings of the renal parenchyma, and amniotic fluid volume have proved unreliable. ^{[11, 12, 13, 9]}

A retrospective study reviewing different prognostic markers of “poor functioning” kidneys and “good functioning” kidneys on the basis of autopsy or biopsy pathology found that “poor functioning” kidneys had decreased amniotic fluid volume, urine output of less than 2 mL/hr, a urine sodium concentration of more than 100 mEq/L, urine chloride levels of greater than 90 mEq/L, and a urine osmolality level greater than 210 mOsm/L. ^[14]

The first successful fetal decompression operation for hydronephrosis was an open procedure performed in 1981. ^[15] Since then, advances in fetoscopy have eliminated the need for open operations. Current management of LUTO involves US-guided vesicoamniotic shunt placement under local anesthesia. ^[16] Amnioinfusion may be required, in that oligohydramnios tends to limit visualization. In addition, various groups have begun to investigate the role of fetoscopic cystoscopy. ^{[3, 2]}

It has been challenging to determine which patients may benefit from fetal intervention; however, animal models have suggested a potential therapeutic role for in-utero intervention. ^{[1, 17]}

The role of fetal surgery for LUTO was first established after experiments using the fetal lamb, which suggested that resolution of LUTOs improved oligohydramnios and decreased the risk of developing pulmonary hypoplasia. ^{[1, 17]} Whereas initial experiments with bilateral obstruction resulted in fetal demise, obstruction of a single ureter resulted in impaired renal function and histologic changes as compared with an unobstructed kidney. ^[17] Furthermore, obstructed fetal kidneys had a decreased glomerular filtration rate (GFR) and increased loss of sodium and chloride. ^[18]

In later experiments, bilateral obstruction via urethral obstruction and urachal ligation followed by in-utero decompression with suprapubic cutaneous cystostomy resulted in improved renal function with fewer histologic changes and more favorable pulmonary outcomes. ^[1]

In addition to urine electrolytes and osmolality, additional studies have identified calcium, β₂-microglobulin, and total protein as being of prognostic importance. ^[6]  Repeated fetal urine aspirations, trending osmolarity, and sodium and chloride levels may be of use. Fetal urine laboratory values for selecting patients that may benefit from antenatal therapy are as follows:

• Sodium level of less than 100 mmol/L
• Chloride level of less than 90 mmol/L
• Osmolality level of less than 200 mOsm/L
• Calcium level of less than 8 mg/dL
• β ₂-microglobulin level of less than 6 mg/dL
• Total protein level of less than 20 mg/dL

At present, fetuses with favorable urine electrolyte levels and no indication of renal dysplasia are considered for vesicoamniotic shunting or fetal cystoscopy.

Contraindications for fetal intervention can be either maternal or fetal.

Maternal contraindications include the following:

• Frequent contractions
• Membrane rupture
• Short cervix
• Uncontrolled comorbidities that may predispose the patient to preeclampsia or HELLP (Hemolysis, Elevated Liver enzymes, Low Platelet count) syndrome

Fetal contraindications include the following:

• Multiple fetal anomalies
• Chromosomal abnormalities
• Anatomic restrictions, including an anterior-lying placenta, which may prohibit safe access

Best practices

Currently, fetal intervention plays no role in the management of UUTOs. An algorithm for managing patients with LUTOs is illustrated in the image below. Because the role of fetoscopic cystoscopy is not yet well defined, it is not included. 

Treatment algorithm for fetal lower urinary tract obstructions. From The Unborn Patient: The Art and Science of Fetal Therapy (p. 272), by Johnson MP, Philadelphia: W.B. Saunders Company. 2001. Adapted with permission.

View Media Gallery

Antenatal US should first be performed to screen for associated anomalies. If no other anomalies are present, fetal karyotyping is recommended. If the patient has a normal male karyotype, the next step is to obtain serial fetal urine samples. If serial electrolyte and β₂-microglobulin levels and the total protein profile are favorable, vesicoamniotic shunt placement is recommended.

Deviation from this pathway portends a poor prognosis, and counseling regarding expectations is currently the best practice.

In 2017, Ruano et al proposed a staging system to guide best practice. Stage I is characterized by a normal amniotic fluid index, no renal echogenicity or cortical cysts, and a favorable urinary biochemical profile; antenatal surveillance is warranted. Stage IV is suggested by in-utero renal failure, as defined by anhydramnios and renal dysplasia, and is not survivable. Stages II and III fall between these extremes and are the stages in which vesicoamniotic shunt placement plays a role. ^{[19, 16]} Further validation studies are needed, but this proposed staging system offers an opportunity for standardization. 

Procedural planning

Antenatal US is an essential component of the LUTO algorithm; anhydramnios poses a challenge to vesicoamniotic shunt deployment, in that the catheter requires a fluid pocket for coiling. Therefore, amnioinfusion is frequently required in severe cases.

The mortality associated with LUTO ranges from 33% to 75%, depending on coexisting anomalies and the severity of oligohydramnios. ^[20] The presence of early oligohydramnios carries a mortality in the range of 45-80%, ^{[21, 22]} probablyresulting from pulmonary hypoplasia. Most fetuses who have an antenatal diagnosis of LUTO with a normal volume of amniotic fluid at midgestation will have a favorable outcome in terms of perinatal survival and are unlikely to need long-term respiratory support, though they remain at increased risk for renal disease. ^[23]  

Outcome analysis suggests that 25-30% of patients diagnosed with LUTOs will require dialysis or renal transplantation. ^{[22, 24]}  There appears to be a perinatal advantage to in-utero vesicoamniotic shunt placement in selected patients, but more studies are needed to determine the long-term benefit. ^{[25, 26, 27]}

To date, no randomized controlled trials have compared the success rates of antenatal and postnatal therapy for LUTO. In 1997, a retrospective analysis of 169 successful percutaneous vesicoamniotic shunts placed over a 14-year period calculated the overall survival to be 47%, and 40% of survivors developed end-stage renal disease. ^[28]

A benefit of fetal cystoscopy as compared with vesicoamniotic shunting is that it can be used as a tool for diagnosing the underlying etiology of LUTO. ^[29] Reported survival rates are 40% for vesicoamniotic shunting and 75% for fetal cystoscopy. Ruano found that postnatal normal renal function was evident in 50% of those who previously underwent vesicoamniotic shunt placement versus 65% of those who underwent fetal cystoscopy. ^[30]  

In a subsequent retrospective study, however, Vinit et al found no differences in survival or rates of chonic kidney disease between patients undergoing vesicoanmniotic shunting and those undergoing fetal cystoscopy. ^[31] Further studies are needed to elucidate potential differences in outcomes between vesicoamniotic shunt placement and fetal cystoscopy.