Pronostic rénal à long terme des enfants présentant une obstruction sous-vésicale congénitale

25 octobre 2018

Auteurs : N. Berte, I. Vrillon, O. Larmure, V. Gomola, C. Ayav, C. Mazeaud, J.L. Lemelle
Référence : Prog Urol, 2018, 12, 28, 596-602




 




Introduction


Congenital Lower Urinary Tract Obstructions (LUTO) is a heterogeneous group of diseases in which urine elimination is obstructed at the level of bladder neck or urethra. This obstruction in utero leads to a global injury of the urinary tract. The mechanisms of such injuries are well known, associating dilatation and fibrosis of bladder and ureter and bilateral renal hypoplasia or dysplasia [1, 2, 3, 4]. Congenital LUTO are the most severe group of Congenital Anomalies of the Kidney and the Urinary Tract (CAKUT). CAKUT are the first cause of pediatric Chronic Renal Disease (CRD) and End Stage Renal Disease (ESRD) [5, 6, 7, 8].


Aetiologies of congenital LUTO are numerous and heterogeneous. The prevalence of LUTO is reported as 2.24 live births per 10,000 live births [9]. The most common etiology is Posterior Urethral Valves (PUVs), affecting only male fetuses and representing 63% of congenital LUTO. Other aetiologies of LUTO are urethral atresia, urethral stenosis, megalourethra, anterior urethral valves, Prune-Belly Syndrome, cloacal anomalies, obstructive ureterocele and severe congenital bladder-sphincter dysfunction [9].


Diagnosis of LUTO must be confirmed in emergency. The treatment is based on urinary diversion and/or obstacle ablation. Bladder catheterization, suprapubic stent, vesicostomy, ureterostomy or nephrostomy are options [10]. In PUVs, primary endoscopic valve ablation in the first days of life without bladder catheterization is the gold standard, permitting to better preserve bladder function [11]. Anaesthetic risk and surgical trauma must be considered in fragile patients for whom bladder catheterization becomes an option.


Long-term outcome of patients suffering of congenital LUTO is marked by renal impairment who determines the prognosis. In patients with congenital LUTO treated with Vesico Amniotic Shunt (VAS), the ESRD rate varies from 33% to 36% and the CRD rate is 22% [12, 13]. In patients with PUVs, the incidence of ESRD is 10% at age 10 years and 38% at age 20 years, with highest incidence occurring in the first year of life and in late adolescence. The incidence of chronic renal failure is 34% at age 10 years, and 51% at age 20 years [11]. A review showed an ESRD rate of 11% (0-20%) and a CRD rate of 22% (0-32%) in patients with PUVs treated by endoscopic valve resection. In this study, elevated nadir serum creatinine was the only independent factor associated with renal failure [14]. The other aetiologies of LUTO are also responsible of renal impairment [15, 16].


The aim of the study is to evaluate long-term renal outcome of patients suffering of congenital LUTO. In order to precise this evaluation, we will compare the long-term outcome between two groups of patients: the patients suffering of PUVs in one hand, and the patients suffering of congenital LUTO of other aetiologies in another hand.


Patients and methods


We retrospectively reviewed the database of the Children's Hospital for cases of congenital LUTO followed-up by the service of Paediatric Nephrology between 1989 and 2012 with at least 1 year follow-up. A total of 40 cases were identified and separated in two groups. Diagnosis was based on clinical examination, surgical exploration including urethral endoscopy or open surgery findings, and imaging such as voiding cystourethrogram.


Data on surgery were obtained for each patient, including the date and the type of surgery.


Data to assess renal outcome were collected at neonatal period and all along the follow-up. During the first year of life, we collected all serum creatinine values, permitting to determinate the nadir serum creatinine value. During the follow-up, serum creatinine values were collected at least once a year, outside of acute renal failure. Schwartz estimate creatinine clearance (glomerular filtration rate) was calculated using the revised Schwartz method. In patients older than one year, Chronic Renal Disease (CRD) was defined as creatinine clearance between 60mL/min/1.73m2 and 15mL/min/1.73m2 without need for initiation of kidney replacement therapy (dialysis or transplantation). These values correspond to CRD stages 3 and 4. End Stage Renal Disease (ESRD) was defined as creatinine clearance below 15mL/min/1.73m2 or need for initiation of kidney replacement therapy, corresponding to CRD stage 5.


Data on renal imaging were also collected. Uretero-Hydronephrosis (UHN) was quoted if significant dilatation of pelvis or calices was detected on ultrasound on a functioning kidney. These UHN were classified as unilateral UHN if affecting only one of two functioning kidneys, and bilateral UHN if affecting two functioning kidneys or one solitary functioning kidney. Renal parenchymal aspect in ultrasonography was also analysed and quoted in hyperechogenic cortex, loss of parenchymal differentiation and cortical microcysts. Differential renal function was evaluated with radioisotope renography such as DMSA scan or MAG3 scan. Asymmetry was quoted if one of the functioning kidneys had a relative function under 40% of the total renal function. Atrophy was quoted when the relative function of a kidney was below 5% or when the renal parenchyma was invisible in ultrasonography.


Finally, bladder function was assessed: the use of urinary drainage for a long-term period such as nephrostomy, cystostomy or trans-urethral intermittent bladder catheterization was quoted. In absence of urinary derivation, the presence of a nighttime incontinence, daytime incontinence or permanent incontinence was quoted in patients older than 5 years.


Statistical analysis was performed to compare the two groups of patients using a statistical software SAS v9.3 (SAS Institute Inc., Cary, NC, USA), and interpreted with the Epidemiological and Clinical Evaluation Service of Nancy Hospital. Chi2 test was used for qualitative variables, and Wilcoxon test was used for quantitative variables. A P -value less than 0.05 was considered significant. Independent t-test was used to compare the means of the two groups, with a confidence interval of 95%. Follow-up varied significantly among patients. A Kaplan-Meier method was used to calculate the incidence of CRD. Because of an initial improvement of the renal function during the first year of life, corresponding to the consequence of the surgical management, glomerular filtration rate was included in the Kaplan-Meier curve after the first year of life. The ESRD incidence was not calculated with Kaplan-Meier method because the number of events was too small.


Results


Forty patients with congenital LUTO were included, 23 patients with PUVs (57.5%) and 17 patients suffering of congenital LUTO of other aetiologies (42.5%): these diagnoses were urethral atresia (2 patients), anterior urethral valve (2 patients), urethral stenosis (2 patients) and cloacal malformation (2 patients). Seven patients presented severe neonatal bladder-sphincter dysfunction with anatomical anomalies of urethra and bladder neck such as bladder neck hypertrophy, without neurological abnormalities. Two patients had other anomalies: a bladder trigone malformation with bilateral ectopic ureteral implantation for one of them, and an obstructive ureterocele for the other one. All patients with PUVs were male, 6 patients with other aetiologies were female, representing 35% of this group and 15% of the whole study. Mean follow-up time was 8.2 years for both groups (range 1 to 21 years), 7.6 years for PUVs (range 2 to 21 years) and 8.8 years for other aetiologies (range 2 to 17 years).


Surgical management


Surgical treatment of obstructive uropathy was performed in 55% within the first day of life (n =22), more frequently in PUVs (60%) than in other aetiologies (47%), without statistical significance. In patients with delayed diagnosis, a statistical difference between the mean delays was assessed between the two groups, with longer delay before surgical management in other aetiologies (P =0.035). First surgical treatment was endoscopic valve resection in 57% of the cases (n =23), and urinary diversion in the other cases (n =17). The mean age at first surgery was 12 days (0.4 month).


A total of 119 surgical operations have been performed in the patients of the study, 45.5% in PUVs (n =90) and 54.5% in other aetiologies (n =108). The patients with LUTO of other aetiologies had proportionally higher number of surgical interventions (6.4 operations per patient) than patients with PUVs (3.9 operations per patient) (P =0.014).


In patients with PUVs, initial surgical treatment consisted in endoscopic valve resection in 43% of the patients (n =10), suprapubic stent followed by endoscopic valve resection within one week in 40% (n =9), and urinary diversion in 17% (n =4).


Mini-invasive surgery, such as endoscopic valve resection, placement of a suprapubic stent or percutaneous nephrostomy were performed statistically more frequently in patients with PUVs patients (67.8% of the total surgical operations) than in patients with other aetiologies (41.7%) (P <0.01). Urological operations represented 98.9% of the operations for PUVs patients, while they represented only 83.3% for other aetiologies, showing statistical difference between the two groups (P <0.01). Patients with LUTO of other aetiologies were more frequently operated for visceral surgery than patients with PUVs (P <0.01).


Renal imaging


Ultrasonographic aspect of the kidneys was frequently altered with 34 bilateral hydronephrosis (85%) and 19 loss of parenchymal differentiation (47.5%). Radioisotope renography showed atrophy in 18 patients (45%) and asymmetry in 16 patients (40%). No significant differences were assessed between the two groups. Table 1 shows the results concerning imaging in each group (Table 1).


Renal outcome


The mean nadir serum creatinine was 5.4mg/L (standard deviation 2.4mg/L) in the patients of the study, without significant difference between PUVs (5.7mg/L, SD 2.9mg/L) and other aetiologies (4.9mg/L, SD 1.3mg/L).


In congenital LUTO (both groups), CRD rate was 17% at age 5 years, 37% at age 10 years and 57% at age 15 years (Figure 1). At the end of follow-up, CRD rate was 25% (10 patients) and ESRD rate was 7.5% (3 patients). Two patients underwent renal transplantation and one was treated by hemodialysis.


Figure 1
Figure 1. 

Incidence of CRD in congenital LUTO.




In PUVs, CRD rate was 20% at age 5 years, 42% at age 10 years and 68% at age 15 years (Figure 2). At the end of follow-up, CRD rate was 26% (6 patients) and ESRD rate was 9% (2 patients). Patients with ESRD underwent renal transplantation.


Figure 2
Figure 2. 

Comparison of incidence of CRD between groups 1 and 2.




In LUTO of other aetiologies, CRD rate was 15% at age 5 years, 30% at age 10 years and 44% at age 15 years (Figure 2). At the end of follow-up, CRD rate was 24% (4 patient) and ESRD rate was 6% (1 patient). The patient with ESRD was treated by hemodialysis.


There is significant statistical difference between incidence of CRD in the two groups (P =0.032), renal prognosis of PUVs being worst that the one of other aetiologies of LUTO.


At the end of follow-up, 8 patients (20%) with congenital LUTO (both groups) used urinary derivation to evacuate urine. Only 2 patients with PUVs (8.6%) used this method, while 6 patients (35.3%) used it in LUTO of other aetiologies. There is significant statistical difference between the two groups (P =0.038).


Prenatal diagnosis was performed in 29 patients (73%) and was not associated with a worsening of renal prognosis, even in case of early prenatal diagnosis.


Presence of retentional bladder wall changes and/or bladder neck hypertrophy during cystoscopy was assessed in 8 patients and was significantly associated with higher risk of ESRD (P =0.036, relative risk=8).


Description of parenchymal loss of renal differentiation during the first ultrasonography was significantly associated with an increasing risk of ESRD at the end of follow-up (P =0.014). In the same way, cortical microcysts in first ultrasonography were associated with poor renal outcome and higher risk of ESRD (P <0.01). However, bilateral hydronephrosis and hyperechogenic cortex were not significantly associated with worst renal outcome.


Data on continence were obtained on 20 patients of the 27 patients older than 5 years of the study (74%). On these patients, 5 patients had urostomy (25%), 3 used intermittent catheterization (15%), 5 had incontinence (25%; 1 daytime, 2 nighttime, 2 permanent), and 7 were dry (35%). Renal prognosis was not significantly different between continent and incontinent patients.


Discussion


The population of our study seems to be similar to other populations of patients with congenital LUTO studied in the literature. One of the main studies on congenital LUTO was done by Malin et al. [9] and reported 284 LUTO cases. The Table 2 compares the aetiologies and their frequency between this study and the present study and shows similar rates (Table 2). Difficulty to classify some rare anatomical anomalies is assessed by a high rate of LUTO unspecified.


Patients of both groups have the same type of early surgical management within the first day of life. However, when the diagnosis is delayed, the delay is longer in patients with LUTO of other aetiologies. This extended delay could be explained by the difficulty to diagnose obstruction in absence of clear urinary retention signs and by the anatomical complexity of these patients. Impact of this postponed diagnosis is difficult to evaluate, but Bomalaski et al. showed that ESRD could occur even in patients with delayed presentation of PUVs [17].


The management of PUVs by endoscopic valve resection is known to be the best treatment [11] and has been performed alone or associated with transitory suprapubic stent in 83% of patients with PUVs. This rate is in the same range than Smith et al. study which showed a rate of 74% of endoscopic valve resection in a series of 100 boys [11]. However, this mini-invasive treatment can rarely be performed in patients suffering of congenital LUTO of other aetiologies for different reasons (anatomical abnormality too complex to explore by endoscopy, level of obstruction impossible to assess during endoscopy, major dilatation of upper urinary tract requiring a high urinary diversion prior to management of the cause of the obstruction). This increases the number of operations compared to patients with PUVs during the follow-up. These data converge with Smith et al. study [11], which showed a significantly higher number of procedures in patients with PUVs who underwent primary diversion. Finally, the high rate of complex associated anatomical anomalies in patients with congenital LUTO of other aetiologies tend to complicate surgical management all along the follow-up, associating complex visceral and genital surgery. The investigation of such associated anomalies during antenatal period is crucial, because they tend to darken the prognosis of these patients.


Concerning the evaluation of renal outcome, incidence of CRD calculated with Kaplan-Meier method has our preference because of the important variation of follow-up time. Our study shows that patients with PUVs have a significantly worst renal prognosis than patients with LUTO of other aetiologies (P =0.03). At the end of the follow up, our study shows a statistically lower rate of ESRD than other series of congenital LUTO. The ESRD rate is 7.5% in our study, 36% in Freedman et al. study (P =0.004) [12] and 33% in Biard et al. study (P =0.008) [13]. This difference can be explained by the selection of the patients because these two studies interested with patients who received an antenatal treatment for LUTO and had antenatal criterions of bad prognosis, which is not the case of our patients. That is why the renal outcome of our series cannot be compared to the one of congenital LUTO series in the literature. Concerning patients with PUVs, the incidence of CRD in 10 years old patients are in the same range in our study (42%) and in Smith et al. study (34%). Renal outcome of patients with PUVs of our study seems to be similar to the one described in literature.


Three parameters observed in the beginning of the follow-up were associated with poor renal outcome in the future. The first is retentional bladder wall changes and/or bladder neck hypertrophy observed during cystoscopy. This cystoscopy being frequently required during the management of these uropathy, attentive observation of the bladder neck and of trabeculations of the bladder wall should be performed. The two other parameters are loss of renal differentiation and cortical microcysts during the first renal ultrasonography. A renal ultrasonography performed by experienced radiologist seems to be mandatory from the beginning of the management of these difficult patients.


Conclusion


Our study shows that long-term renal outcome of patients with congenital LUTO is associated with high incidence of Chronic Renal Disease. Renal prognosis of patients with Posterior Urethral Valves is significantly worse than patients with other aetiologies of congenital LUTO. However, the anatomical complexity of patients suffering of congenital LUTO other than PUVs is responsible of a higher use of urinary diversion, and a more invasive surgical management. From the beginning of management, existence of bladder neck hypertrophy or retentional bladder wall changes in cystoscopy, and cortical microcysts or parenchymal loss of differentiation in ultrasonography are significantly associated with higher risk of renal failure in the future. These informations are useful when facing such rare congenital urological anomalies to better predict future prognosis of these patients.


Disclosure of interest


The authors have not supplied their declaration of competing interest.




Table 1 - Results of renal imaging in each group.
  Congenital LUTO (Total) 
PUVs 
Other aetiologies 
P ** 
  n =40 
n =23 (57.5%) 
n =17 (42.5%) 
 
  n   n   n    
Uretero Hydro Nephrosis              0.6556 
No  7.5  8.7  5.9   
Unilateral UHN  7.5  4.3  11.8   
Bilateral UHN  34  85  20  87  14  82.4   
Loss of parenchymal differentiation              0.2176 
Yes  19  47.5  39.1  10  58.8   
No  21  52.5  14  60.9  41.2   
Hyperechogenic cortex              0.8535 
Yes  30  75  17  73.9  13  76.5   
No  10  25  26.1  23.5   
Cortical microcysts              0.6869 
Yes  15  13  17.6   
No  34  85  20  87  14  82.4   
Asymmetry (<40%)              0.8961 
Yes  16  40  39.1  41.2   
No  24  60  14  60.9  10  58.8   
Atrophy (<5%)              0.0884 
Yes  18  45  13  56.5  29.4   
No  22  55  10  43.5  12  70.6   





Table 2 - Comparison of LUTO etiologies between present study and Malin et al. study.
LUTO subtype  Malin et al. 
Present study 
  Number  % LUTO  Number  % LUTO 
Posterior urethral valves  179  63.0  23  57.5 
Urethral atresia  28  9.9  5.0 
Urethral stenosis  20  7.0  5.0 
Prune Belly syndrome  2.5  0.0 
LUTO other  0.7  15.0 
LUTO unspecified  48  16.9  17.5 
Total LUTO cases  284  100.0  40  100.0 




References



Truong L.D., Gaber L., Eknoyan G. Obstructive uropathy Contrib Nephrol 2011 ;  169 : 311-326 [cross-ref]
Chevalier R.L., Forbes M.S., Thornhill B.A. Ureteral obstruction as a model of renal interstitial fibrosis and obstructive nephropathy Kidney Int 2009 ;  75 (11) : 1145-1152 [cross-ref]
Thomas J. Etiopathogenesis and management of bladder dysfunction in patients with posterior urethral valves Indian J Urol 2010 ;  26 (4) : 480-489 [cross-ref]
Casella D.P., Tomaszewski J.J., Ost M.C. Posterior urethral valves: renal failure and prenatal treatment Int J Nephrol [Internet] 2012 ; [cité 21 févr 2013].
PMC3154780/
Ardissino G., Daccò V., Testa S., Bonaudo R., Claris-Appiani A., Taioli E., et al. Epidemiology of chronic renal failure in children: data from the ItalKid project Pediatrics 2003 ;  111 (4 Pt 1) : e382-e387
Hattori S., Yosioka K., Honda M., Ito H. The 1998 report of the Japanese National Registry data on pediatric end-stage renal disease patients Pediatr Nephrol 2002 ;  17 (6) : 456-461 [cross-ref]
Lewis M.A., Shaw J., Sinha M., Adalat S., Hussain F., Inward C. UK Renal Registry 11th Annual Report (December 2008): Chapter 13 Demography of the UK paediatric renal replacement therapy population Nephron Clin Pract 2009 ;  111 (Suppl. 1) : c257-c267
Yosypiv I.V. Congenital Anomalies of the Kidney and Urinary Tract: A Genetic Disorder? Int J Nephrol [Internet] 2012 ; [cité 21 févr 2013].
PMC3363415/
Malin G., Tonks A., Morris R., Gardosi J., Kilby M. Congenital lower urinary tract obstruction: a population-based epidemiological study BJOG Int J Obstet Gynaecol 2012 ;  119 (12) : 1455-1464 [cross-ref]
Séminaire de chirurgie pédiatrique viscérale, Audry G, Société française de chirurgie pédiatrique, Section française d'urologie pédiatrique. Les uropathies congénitales de diagnostic anténatal  : Sauramps médical (2011). 
Smith G.H., Canning D.A., Schulman S.L., Snyder H.M., Duckett J.W. The long-term outcome of posterior urethral valves treated with primary valve ablation and observation J Urol 1996 ;  155 (5) : 1730-1734 [cross-ref]
Freedman A.L., Johnson M.P., Smith C.A., Gonzalez R., Evans M.I. Long-term outcome in children after antenatal intervention for obstructive uropathies Lancet 1999 ;  354 (9176) : 374-377 [cross-ref]
Biard J.-M., Johnson M.P., Carr M.C., Wilson R.D., Hedrick H.L., Pavlock C., et al. Long-term outcomes in children treated by prenatal vesicoamniotic shunting for lower urinary tract obstruction Obstet Gynecol 2005 ;  106 (3) : 503-508 [cross-ref]
Hennus P.M.L., van der Heijden G.J.M.G., Bosch J.L.H.R., de Jong T.P.V.M., de Kort L.M.O. A systematic review on renal and bladder dysfunction after endoscopic treatment of infravesical obstruction in boys PLoS One [Internet] 2012 ;  7 (9) : [quoted february 2017].
PMC3441498/
Warne S.A., Wilcox D.T., Ledermann S.E., Ransley P.G. Renal outcome in patients with cloaca J Urol 2002 ;  167 (6) : 2548-2551[discussion 2551].
 [cross-ref]
Amsalem H., Fitzgerald B., Keating S., Ryan G., Keunen J., Pippi Salle J.L., et al. Congenital megalourethra: prenatal diagnosis and postnatal/autopsy findings in 10 cases Ultrasound Obstet Gynecol 2011 ;  37 (6) : 678-683 [cross-ref]
Bomalaski M.D., Anema J.G., Coplen D.E., Koo H.P., Rozanski T., Bloom D.A. Delayed presentation of posterior urethral valves: a not so benign condition J Urol 1999 ;  162 (6) : 2130-2132 [cross-ref]






© 2018 
Elsevier Masson SAS. Tous droits réservés.