From prenatal LUTO intervention to postnatal surgery: complication rates and operative requirements

Introduction

Fetal lower urinary tract obstruction (LUTO) is defined as a group of diseases characterized by a more or less pronounced obstruction of the outflow from the lower urinary tract. A higher prevalence of the condition has been observed in male fetuses, and the most common cause is posterior urethral valves [1]. The condition can also result from various other congenital anomalies, including anterior urethral valves, urethral hypoplasia, stenosis or atresia, anorectal malformations, and other syndromic diseases. Prenatal urine retention due to outflow obstruction can lead to the development of megacystis, hydronephrosis, vesicoureteral reflux, renal dysplasia, and subsequent kidney damage [2], 3]. Furthermore, a reduction in amniotic fluid (oligohydramnios) resulting from urine obstruction can lead to complications in lung development, such as pulmonary hypoplasia, as well as limb deformities [2]. An early diagnosis is typically made by prenatal ultrasound, which can reveal abnormalities such as megacystis or hydronephrosis [4].

Prenatal interventions, including vesicoamniotic shunting (VAS) and amnioinfusion, are implemented with the objective of restoring urine flow and minimizing renal and pulmonary complications. In the extant literature, the efficacy, optimal timing, and selection criteria for prenatal intervention remain under debate, while there is a paucity of data addressing complications that necessitate postnatal surgery. The present study investigates the postnatal management of complications caused by prenatal interventions in children with LUTO.

Materials and methods

This retrospective study analyzes the data of male children who were treated postnatal for LUTO between January 1st, 2013, and December 31st, 2022, at the department of pediatric surgery and pediatric urology of the University Hospital of Marburg. The inclusion criterion for this study was the prenatal intervention for LUTO. All patients who were not treated at Marburg University Hospital immediately after birth and whose postnatal treatment record was incomplete were excluded. As the patients were initially referred following prenatal intervention in a center specialized in prenatal medicine, no information is available regarding the indication for intervention. All prenatal treatments were carried out at two different centers. Prenatal intervention included VAS placement or/and an amnioinfusion. Data were manually extracted from patient records, including prenatal diagnoses, gestational age at diagnosis, timing and frequency of prenatal intervention, and postnatal development including pediatric surgical intervention. The observation period extended from 2013 to 2024. The ethics committee of the University Hospital Marburg has given a positive vote for the study (ethics committee university of Marburg, Germany, 06.12.2022, RS 22/63).

Results

A total of 23 patients with LUTO and prenatal intervention were included in the study. The diagnosis of LUTO varied between the 12th and 26th week of gestation (WG). On average, the diagnosis was made in the 17.1th WG (standard deviation (SD) of ±7).

On prenatal ultrasound, a keyhole sign was observed in 12 of 23 cases (55 %), megacystis in 17 of 23 fetuses (77 %), hydronephrosis in 11/23 children (50 %), and at least one dilated ureter in 10 children (46 %). Oligohydramnios or anhydramnios was diagnosed in 10 children (43 %) peri-interventional and in 19 patients (83 %) at the last prenatal ultrasound.

Stent-associated complications

In 4 children (17 %), a fistula occurred, 2 vesicoperitoneal and 2 vesicorectal fistulas. In both vesicoperitoneal fistulas, the bladder wall was reconstructed. The vesicorectal fistulas were excised after removal of the inserted stents, and both the bladder wall and the rectal wall were subsequently reconstructed. All fistulas could be closed primarily without complications.

In 5 children (21 %), an abdominal wall defect appeared. The abdominal wall defects ranged from a few millimeters to several centimeters in size. In 2 patients, relevant parts of the small intestine were prolapsed through the abdominal wall defect (Figures 1 and 2). In the smaller defects, the greater omentum was eventuated. All abdominal wall defects were reconstructed primarily.

Figure 1:

Newborn with intestinal prolapse through abdominal wall defect after prenatal VAS.

Figure 2:

Newborn with abdominal wall defect as well as jejunal atresia after prenatal VAS.

Renal perforation, ureteral or urethral injury, occurred each in 1 patient (4 %) (Figures 3 and 4). The renal stent was retrieved by opening the renal pelvis. The renal perforation and the renal pelvis were then reconstructed. The ureteral perforation was reconstructed primary. The stent that injured the urethra has been placed through the bladder neck into the prostate. The stent was removed transvesically, and the urethral perforation healed conservatively after placement of a bladder catheter. In one patient, the T-anchor was left in the renal parenchyma because removal would have caused considerable damage.

Figure 3:

Intra-abdominal dislocation of VAS, abdomino-amniotic shunt, as well as t-anchor.

Figure 4:

Postnatal retrograde contrast study of the bladder: evidence of vesicoureteral reflux and perforation of the renal pelvis.

Bowel perforation occurred in 3 cases (12 %), one jejunal, one ileum, and one colon transversum. The jejunal perforation led to atresia and intestinal loop prolapsing in front of the abdominal wall. Intestinal continuity was reconstructed and the abdominal wall closed. In the case of ileum perforation, a segment was resected and the bowel was reconstructed. A transversostomy was placed in the unstable newborn with colon perforation.

The following therapeutic measures were taken regarding the cause of the LUTO. All patients with urethral valves underwent laser ablation of the valves without prior vesicostomy, requiring 1–3 laser ablations. In cases of urethral stenosis, patients initially underwent vesicostomy, followed by progressive augmentation by dilation of the anterior urethra (PADUA). The PADUA procedure was successful in two out of four patients. The remaining two patients and those diagnosed with urethral atresia required vesicostomy in the long term.

According to the neonatal documentation, a diagnosis of pulmonary hypoplasia was made in 10 cases. Thirteen patients required mechanical ventilation postnatally for a maximum period of 53 days. The patient did not require extracorporeal membrane oxygenation. All patients were discharged without the need for respiratory support.

In the follow-up, 21/23 patients had chronic renal failure. In the present study, no discrepancy in renal function was observed between the early and late intervention groups. Four of the five patients with a VAS before 15 WG developed chronic kidney disease. Thirteen children (57 %) required peritoneal dialysis, with 4 later switching to hemodialysis due to complications. The median dialysis initiation occurred after 71 days of life (range: 1 day to 32 months). Five children (22 %) received kidney transplants. Two patients died, with one patient dying immediately postnatally due to pronounced pulmonary hypoplasia. The other patient died at the age of 9 months due to pulmonary embolism. The average follow-up period was 36.3 months (2–113 months). The results are represented in Table 1.

Table 1:

Summary of patients with LUTO and prenatal intervention.

Patient	GA – diagnosis	Amniotic fluid	GA – 1. VAS	Amnioinfusion	Additional VAS	Additional shunts	GA – birth	Diagnosis	Shunt complication	Shunt dislocation	Pulmonary hypoplasia	Kidney functiona	Time to dialysis (d)	Death
1	13	OH	14 + 1	No	2		40	US		Yes	No	Normal
2	25	OH	27 + 5	No	1	Uretero-amniotic + abdomino-amniotic	35	PUV		Yes	No	Dialysis	122
3	18	OH	18 + 5	No	1	Reno-amniotic + abdomino-amniotic	32	PUV	Perforation of the kidney and renal pelvis	Yes	No	CKD
4	15	AH	15 + 3	Repeated	0		37	PUV	Abdominal wall defect	Yes	Yes	KTx	399
5	15	AH	18 + 5	Repeated	0		34	PUV	Abdominal wall defect	Yes	Yes	Dialysis	11
6	19	OH	20 + 4	No	1	Uretero-amniotic + abdomino-amniotic	35	PUV	Perforation of ureter and urethra	Yes	No	CKD
7	16	AH	16 + 1	Repeated	1	Uretero-amniotic	33	PUV	Ileum perforation	No	Yes	Dialysis	63	290 d
8	13	AH	14 + 3	Repeated	0		35	PUV		No	No	Dialysis	1,100
9	13	AH	14 + 3	No	0	Uretero-amniotic + abdomino-amniotic	35	PUV		Yes	Yes	Dialysis	32
10	16	AH	17 + 2	No	0		37	UA		No	No	KTx	174
11		AH		No	2		33	US	Abdominal wall defect	No	No	CKD
12d	19	AH	20 + 0	No	1	Abdomino-amniotic	38	AUV		Yes	Yes	KTx	684
13	15	Normal	14 + 6	No	0		36	PUV	Abdominal wall defect, jejunum perforation/atresia	Yes	No	CKD
14		OH	14 + 2	Once	1		37	UA		Yes	No	CKD
15	19	Normal	28 + 5	Once	0		34	PUV		No	No	CKD
16	21	OH	21 + 6	Onceb	0		37	PUV		Yes	No	CKD
17c	12	AH	16 + 5	No	1		30	US	Abdominal wall defect, colon perforation	Yes	Yes	Dialysis	1
18	16	Normal	18 + 0	No	0		38	UA		No	No	CKD
19	13	AH	16 + 0	Repeated	0		36	PUV		No	Yes	KTx	1,003
20	21	AH	21 + 6	Repeated	2		37	PUV		Yes	No	Dialysis	71
21		AH	Unkown	Repeated	0	Reno-amniotic	35	US		Yes	Yes	Dialysis	3
22	17	AH	No	Repeated	X		38	PUV		No	Yes	KTx	5
23	18	AH	18 + 4	Repeated	3	Abdomino-amniotic	34	Unknown		Yes	Yes	–		1 d

1. GA, gestational age; VAS, OH, oligohydramnios; AH, anhydramnios; US, urethra stenosis; UA, urethra atresia; PUV, posterior urethral valve; AUV, anterior urethral valve; CKD, chronic kidney disease; KTx, kidney transplantation; d, day, ^aat last follow-up, ^bamnioinfusion prior to VAS, ^cfetal cystoscopic laser ablation of PUV (21 WG), ^dbladder roof was opened fetoscopically (22 WG).

Discussion

The results of this study highlight the challenges encountered postnatally after prenatal intervention for LUTO. The postnatal course depends on the underlying disease and the various pre- and postnatal manifestations of urinary tract obstruction.

Risk assessment plays a key role in prenatal counseling parents based on LUTO and in particular with regard to possible prenatal intervention. While mild megacystis with a longitudinal bladder diameter in the 1st trimester between 7 and 12 mm often regresses spontaneously, restricted urinary bladder drainage is likely with a diameter of >15 mm [5], [6], [7]. The severe forms of LUTO are associated with oligo- or anhydramnios with subsequent pulmonary hypoplasia as well as chronic renal insufficiency, preterm delivery, and increased intrauterine and postnatal mortality [8], [9], [10], [11]. The pre- and postnatal survival rate correlates with the time of prenatal diagnosis as well as with the occurrence of an oligo- or anhydramnios [6], [7], [8]. Diagnosis in the 1st trimester and an oligo- or anhydramnios before the 20th WG are considered risk factors for a negative outcome, and without prenatal intervention, most fetuses die [6], [7], [8].

Prenatal interventions may be performed to improve the prognosis of selected fetuses with LUTO. Accordingly, the current consensus paper of the ERKNet CAKUT-Obstructive Uropathy Work Group recommends prenatal vesicoamniotic shunt treatment in fetuses with moderate and severe LUTO according to the classification of Ruano et al. [7], 12]. The international LUTO Working Group recommends VAS after the 16th WG for fetuses with LUTO and oligo- or anhydramnios [13]. However, a comprehensive meta-analysis encompassing 246 fetuses, with VAS conducted between the 20th and 27th WG, was compared with a conservative approach. This analysis revealed a significant survival advantage in the perinatal period; however, no significant differences in survival rates after 1 or 2 years or in long-term renal function were observed [14].

Despite advancements in prenatal interventions and association with a significant improvement in the probability of perinatal survival [15], the interventions have a considerable risk of pre- and postnatal complications. The risk of complications associated with intervention is stated in meta-analyses and studies with a larger number of cases to be between 22 and 40 % [7], [16], [17], [18], [19]. These complications vary greatly in the different studies and include miscarriage (0–13 %) [18], 20], 21], premature rupture of the membranes (18–30 %) [18], 22], 23], shunt dislocation or migration (3–100 %) [21], 24], 25], shunt occlusion (6–20 %) [23], 26], abdominal wall defects (2–16 %) [18], 25], and intestinal injuries (2 %) [18].

The most common complication, the shunt dislocation or migration, may be influenced by factors such as fetal movement, the anatomical position of the bladder, and the type of shunt used. It could already be reduced by adapting the materials used [17], 19], 27], 28]. The comparison between Somatex and Harrison shunt systems shows a significantly lower dislocation and complication rate with the Somatex shunt [17]. There is a consensus that a new VAS should be considered in the event of shunt dislocation [13]. Accordingly, 12/23 patients in our patient cohort underwent renewed shunt relief of the urogenital system. However, shunt migration through the bladder wall into the abdominal cavity also leads to urinary ascites in a not inconsiderable proportion of fetuses requiring an abdomino-amniotic shunt for relief [18], 21], 25]. In the patient population under investigation, abdomino-amniotic shunting was indicated in six out of 23 fetuses (26 %), a rate that is comparable with the literature.

The placement of reno-amniotic shunts, as performed in two of our patients in addition to VAS, must be viewed critically due to the lack of evidence and the high peri-interventional risk. One of the two patients sustained a renal injury, encompassing the kidney and renal pelvis. It is evident that there is an absence of any quantifiable advantages in terms of preserving renal function or preventing pulmonary hypoplasia.

When counseling parents regarding prenatal intervention, it is imperative to also consider postnatal complications, in addition to those prenatal complications previously discussed. Postnatal surgical interventions were required in almost all infants in our study cohort, which poses an additional risk, especially for premature infants. In a study by Kohaut et al., shunt removal via laparoscopy or laparotomy was necessary in 10/17 patients, and in two patients, a bladder wall reconstruction was indicated [29]. In comparison, laparoscopy or laparotomy was necessary in 18/23 patients in our study.

A significant distinction between our study and extant literature pertains to the prevalence and severity of abdominal wall defects. While the literature documents individual cases of abdominal wall defects involving the prolapsed omentum majus [18], 25], our study reveals a 21 % prevalence of such defects, accompanied by the prolapsing of bowel segments in two cases, resulting in iatrogenic laparoschisis.

In addition to the aforementioned defects, three fetuses exhibited intestinal perforations through the shunt, necessitating postnatal reconstruction of intestinal continuity, a phenomenon that has only been documented in isolated cases in the literature [18]. In addition, the reconstruction of renal perforation, ureteral or urethral injury was necessary in our study.

Understanding these outcomes is essential for guiding clinical decisions and improving the prognosis for affected patients. Moreover, the study’s findings highlight the need for individualized approaches based on each fetus’s unique presentation and the extent of the obstruction. Additionally, complications such as bladder perforation, shunt obstruction, and potential damage to adjacent organs underscore the invasive nature of these procedures.

To potentially improve outcomes, early diagnosis and intervention have been suggested, especially regarding the survival and the preservation of renal function in severe megacities [1], 18], 25]. It is assumed that the long-term damage to the kidney and thus the reduction in kidney function is linked to the duration of the obstruction of the urinary tract [30], 31]. The hypothesis under consideration is that kidney dysplasia has not yet developed prior to the 16th WG [30], 31]. The initiation of renal function, from the 11th to the 13th WG, and the contribution of renal excretion to amniotic fluid levels, from the 16th to the 18th WG, are crucial factors in this regard [32]. Furthermore, in fetuses with a preterm rupture of membranes, it was shown that even a relatively short duration of oligohydramnios leads to pulmonary dysplasia [33]. Postnatal respiratory morbidity correlates with the severity of the prenatal oligohydramnios at diagnosis, whereby the cause of the oligohydramnios is of secondary importance [34], 35]. Accordingly, contrary to the recommendations of the ERKNet CAKUT-Obstructive Uropathy Work Group and the international LUTO Working Group, some working groups recommend intervention before the 16th WG [1], 18], 25], 36].

However, earlier interventions, especially those before the 20th WG, may carry higher risks of complications. Consequently, the efficacy of early intervention is also the subject of critical debate. In our study, six children received interventions before 16 WG, and five of these experienced complications such as shunt dislocation, abdominal wall defects, or jejunal atresia. Comparable complications and in particular the need for repeated shunt application were also described by Debska et al. [25]. All 6 fetuses included in her study required a new VAS due to a dislocation and a severe oligohydramnios. The increased risk of shunt dislocation in early intervention is attributed to a number of factors, including rapid fetal growth, increased movement, and a thinner bladder wall [17], 37]. In contrast, a significantly lower complication rate was found in studies by working groups with more experience in early intervention [17], 18]. In the study by Gottschalk et al. the shunt-associated complication rate in the VAS group before 14 WG was even lower than with a later intervention, although prematurity was more frequent and the survival rate was lower [18]. This suggests that while earlier intervention might benefit some patients, the associated risks must be carefully weighed.

Despite prenatal interventions, the postnatal outcomes for children with LUTO in this study were marked by a high incidence of complications. The majority of patients (96 %) developed chronic kidney insufficiency, with 57 % requiring peritoneal dialysis and the need for kidney transplantation in 22 % of patients. In this cohort, prenatal intervention seems not to have improved renal outcome. In the early intervention group, 4 out of 5 children also have chronic renal insufficiency. This points to the severe and progressive nature of the disease in some cases and indicates that prenatal interventions do not always prevent long-term renal damage. This underscores the importance of comprehensive postnatal follow-up, including regular monitoring of renal function, early detection of urinary tract abnormalities, and timely intervention when necessary.

The findings of this study have several implications for clinical practice. Firstly, the decision to perform prenatal interventions should involve careful risk–benefit analysis, considering both the potential for improved renal and pulmonary function and the risk of intervention-related complications. Parental counseling is crucial, as families must be informed about the potential outcomes, including the likelihood of postnatal interventions, the risk of renal failure, and the pre- and postnatal mortality rates. In order to provide parents with comprehensive advice on the pre- and postnatal process, multidisciplinary consultation is essential. Secondly, the high rate of complications associated with shunt placement emphasizes the need for improved techniques and technologies. Future research should focus on developing more stable shunt designs and less invasive methods for decompressing the fetal bladder.

This study’s retrospective nature and relatively small sample size limit the generalizability of the results. Furthermore, prenatal intervention was performed by a total of two centers, and therefore a comprehensive picture of prenatal intervention cannot be provided. The study included only male patients, as LUTO predominantly affects males. Therefore, the findings may not be entirely applicable to the minority of female patients with LUTO. Additionally, due to incomplete prenatal data in some cases, the study could not analyze the impact of different shunt types on outcomes, nor could it explore the potential benefits of early vs. late intervention in a controlled manner.

Conclusions

This study demonstrates that prenatal intervention for LUTO, while potentially lifesaving, is associated with frequent and severe complications, both peri- and postnatally. Most patients experience significant morbidity, including the need for multiple surgeries and an elevated risk of chronic kidney disease. These findings argue for cautious patient selection, comprehensive prenatal counseling, and continued innovation in intervention techniques to improve outcomes for this vulnerable population.