Peritoneal-amniotic shunt in management of urinary ascites complicating fetal lower urinary tract obstruction.

Abstract

Fetal lower urinary tract obstruction (LUTO) is a rare congenital anomaly, resulting from the obstruction of the urethra. Early-onset, prolonged anhydramnios results in pulmonary hypoplasia (PH) and fetal or neonatal demise. Vesicoamniotic shunt (VAS) placement aims to restore amniotic fluid volume (AFV) and promote lung development1. Urinary ascites (UA), resulting from bladder perforation or rupture, can be a temporary complication, resolving within 2 weeks, though occasionally it may persist, limiting the ability to perform VAS. Peritoneal-amniotic shunt (PAS) may be considered in persistent cases as an alternative to VAS. We sought to assess the outcome of cases affected by LUTO with persistent UA and anhydramnios that were treated with PAS. This was a retrospective cohort study of cases with fetal LUTO and persistent UA requiring intervention at our center between January 2013 and January 2020, approved by the institutional review board (H-37494). All patients with LUTO underwent comprehensive ultrasound evaluation (GE Voluson E8 or E10, GE Healthcare, Zipf, Austria). Severe LUTO was diagnosed in the presence of a distended bladder, hydronephrosis and/or hydroureters and anhydramnios. Renal function was determined by morphological appearance of the kidneys on ultrasound, urine biochemical analysis (sodium, calcium, chloride, osmolarity and beta-2-microglobulin) and bladder refill after vesicocentesis2. Cases with multiple gestation, presence of other major fetal structural abnormality or abnormal karyotype were excluded. PAS was offered in cases with LUTO and persistent (≥ 2 weeks) UA. Ultrasound-guided amnioinfusion followed by insertion of a double pigtail catheter (Rocket or Harrison) was performed as previously described3. The choice of shunt type was based on the physician's preference. Baseline demographics, pregnancy and postnatal outcomes and renal function (classified as normal or chronic kidney disease (CKD) based on the Schwartz formula4) at the age of 2 years were analyzed. Of the 33 patients undergoing intervention, two had only PAS, 11 had VAS and PAS and 20 had only VAS. Records were available for 10 of the 13 cases with PAS (Table 1). There were two cases of neonatal demise. Of the 10 cases included in the analysis, seven (six of the eight surviving cases) had UA at the time of the initial shunt placement. Median number of shunts placed, VAS or PAS, was three. Only one PAS was performed in 60% (6/10) of cases, while the remaining patients had two PAS. In 20% (2/10) of cases, further intervention after initial PAS was not required. Median gestational age at initial intervention was 20.9 weeks (range, 16.2–28.3 weeks) and delivery was at 34.2 weeks (range, 26.4–39.0 weeks). Of the eight surviving infants, three had PH requiring prolonged respiratory support and two developed Stage 4 or 5 CKD (Table 2). One case was complicated by ventral hernia with incarcerated bowel at the site of the shunt, which required surgical management. PAS is a feasible treatment option for LUTO with persistent UA, which may improve neonatal survival. In this study, 8/10 cases survived until at least 2 years of age and 60% of patients had only mild-to-moderate renal disease. Shunts have been used in LUTO for continuous drainage of fluid from the bladder to the amniotic cavity, normalizing AFV. Complications associated with shunt placement include shunt blockage and displacement, ascites, preterm labor and chorioamnionitis5. The Harrison shunt has been associated with a higher displacement rate6. UA results from spontaneous rupture of the bladder7, perforation after vesicocentesis8 or VAS9. UA may be temporary, resolving within 2 weeks10; however, if the perforation fails to heal, it limits the ability to place a VAS due to decompression of the fetal bladder and increased risk of deployment of the outer coil of the shunt into the fetal peritoneal space instead of the amniotic cavity. PAS may be considered as an alternative in these cases not only to restore AFV but also to relieve the cardiac compression resulting from severe UA11. Concerns regarding PAS include increased shunt displacement due to decreased abdominal strength, resulting from prolonged distension. Shunt obstruction from bowel has been reported when used for ascites associated with other conditions12. In our series, 20% of cases did not require any further procedures after PAS placement and none had shunt malfunction. Our study is limited by its retrospective nature. An important strength was that all patients had postnatal follow-up by the same team of consultants involved during prenatal evaluation, avoiding delays in diagnosis and treatment and improving outcomes. We believe our experience should help to provide guidance for management of these rare cases of LUTO with atypical presentation and avoid delays in treatment which may worsen outcomes due to the time-sensitive nature of fetal lung development. We would like to thank our amazing pediatric nephrology (Dr Michael Braun and Dr Joseph Angelo) and urology (Dr Chester Koh and Dr David Roth) teams which have been an essential part of the counseling and management of all our patients. The data that support the findings of this study are available from the corresponding author upon reasonable request.