Enterocystoplasty in children with neuropathic bladders: Long-term follow-up

Journal of Pediatric Urology (2008) 4, 27e31

Enterocystoplasty in children with neuropathic bladders: Long-term follow-up ´pez Pereira*, J.A. Moreno Valle, L. Espinosa, P. Lo J.M. Alonso Dorrego, L. Burgos Lucena, M.J. Martı´nez Urrutia, R. Lobato Romera, M. Luz Picazo, J.M. Viguer, E. Jaureguı´zar Monereo Department of Pediatric Urology, University Hospital La Paz, P Castellana 261, Madrid 28046, Spain Received 27 June 2007; accepted 30 July 2007 Available online 17 September 2007

KEYWORDS Bladder augmentation; Enterocystoplasty; Neuropathic bladder

Abstract Objective: This study assesses clinical outcome, after at least 8 years, of augmentation done before or at puberty in neuropathic bladders. Patients and methods: A total of 29 children with neuropathic bladders who did not respond satisfactorily to clean intermittent catheterisation and anti-cholinergic therapy underwent enterocystoplasty at a mean age of 11.8 years (range 3e18). Twenty-one children (72.4%) had vesicoureteral reflux (VUR) and/or ureterohydronephrosis and 22 (75.8%) had dimercapto-succinic acid scars, but all had normal renal function. All patients were followed at regular intervals with urinary tract imaging, serum electrolytes, creatinine, urodynamic evaluation and 24-h urine collection. Urine cytology, cystoscopy and biopsy were performed at the end of follow-up. Results: Mean follow-up was 11 years (range 8e14.5) and mean age at the end of follow-up was 22.2 years (range 13.2e31). Urodynamic studies showed a significant improvement in bladder compliance in all patients. Upper urinary tract dilatation disappeared in all, VUR in 13/17 (76.4%), and no new renal scarring occurred in any patient. At the end of follow-up, renal function was normal in all according to serum creatinine, but cystatin C levels were normal in 27 and elevated in two. Significant proteinuria and low concentrations of renin and aldosterone were present in 80% and 82%, respectively. Only one patient had urinary tract infection, three had bladder stones, and in another a catheterisable channel was made. All patients were dry with normal urine cytology and cystoscopy, and no malignant lesions have been found in the biopsy specimens. Conclusion: Enterocystoplasty has preserved renal function and resolved VUR and/or hydronephrosis in most patients. The future implications of proteinuria and the low serum levels of renin and aldosterone, as well as the best indicator for measuring renal function, have yet to be determined. Close, life-long follow-up, including cystoscopy, is necessary to prevent complications. ª 2007 Journal of Pediatric Urology Company. Published by Elsevier Ltd. All rights reserved.

* Corresponding author. E-mail address: [email protected] (P. Lo ´pez Pereira). 1477-5131/$30 ª 2007 Journal of Pediatric Urology Company. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jpurol.2007.07.009

28

P. Lo ´pez Pereira et al.

Introduction Surgical enlargement of the urinary bladder continues to be the mainstay of reconstruction for those patients with reduced functional bladder capacity and compliance who do not respond to clean intermittent catheterisation (CIC) and/or anti-cholinergic therapy. The purpose of this augmentation is to improve bladder function as a low-pressure reservoir in an attempt to protect the kidneys as well as to achieve continence. Enterocystoplasty is commonly used in children born with developmental abnormalities involving the genitourinary system, and it has become a standard part of neuropathic bladder (NB) management. However, when urinary and intestinal tissues are placed in direct contact numerous complications can occur, including recurrent UTI, spontaneous bladder perforation, urolithiasis, metabolic imbalance and the potential for malignancy. These complications are not insignificant and require lifelong surveillance and follow-up. There are few long-term follow-up reports about patients whose bladders were augmented with intestine during childhood [1e4]. We have assessed the long-term follow-up of neuropathic patients in whom the bladder was augmented with intestine before or at puberty.

Patients and methods We retrospectively reviewed the records of 29 NB patients (18 females, 11 males) who underwent enterocystoplasty before or at puberty and who also had a minimum follow-up of 8 years. The underlying aetiology of NB was myelomeningocele in 27 patients, sacral agenesis in one patient and sacrococcygeal teratoma in one. Before their enterocystoplasty only eight patients had a normal upper urinary tract and 21 patients (72.4%) had VUR and/or ureterohydronephrosis (UHN, Table 1). DMSA scintigraphy showed renal scars in 22 patients (16 unilateral and six bilateral) and in two kidneys the differential renal function was between 18% and 20%. DMSA scintigraphy was only normal in seven patients. Of the 21 patients with VUR and/or UHN, 11 (52%) were referred to us from other centres at different ages. The mean age on starting CIC and anti-cholinergic therapy was 4.4 years (range 8 months to 9 years) and four patients had a prior vesicostomy. All patients had normal renal function (RF) at that time, based on their age-adjusted serum creatinine levels. All patients had at least three standard urodynamic studies before bladder enlargement was considered.

Table 1

Upper urinary tract before enterocystoplasty

Urodynamic evaluation before enterocystoplasty showed a poorly compliant bladder in all patients, associated to detrusor overactivity in six. The mean bladder capacity was 89.8 ml (range 58e252) at a mean detrusor pressure of 44.8 cm of water (range 22e150). In four patients who were incontinent due to low sphincteric resistance we planned to implant an artificial urinary sphincter (AUS) necessitating simultaneous bladder augmentation. Mean bladder capacity in these four patients was 145.7 ml (range 77e188) at a mean detrusor pressure of 28.7 cm of water (range 23e33). Despite CIC and anti-cholinergic therapy, VUR, UHN or bladder compliance did not improve in any patient. The indications for enterocystoplasty in this study are shown in Table 2. In all patients enterocystoplasty was performed according to the technique originally described by Bramble [5] and popularised by Mundy and Stephenson [6]. In those patients with VUR and/or UHN, no anti-reflux procedure was performed, as correction of the poorly compliant high-pressure bladder would theoretically lead to resolution of VUR or UHN. After enterocystoplasty, all patients were placed on an irrigation protocol using saline solution once a day. After CIC, they had to introduce 50e100 ml of saline repeatedly into the bladder three to four times to reduce mucous accumulation and so prevent the formation of bladder stones. Patients were followed at regular intervals with assessment of serum electrolytes, serum creatinine, blood urea nitrogen, venous blood gases, urinary culture, blood pressure, upper and lower urinary tract images, urodynamic evaluation, and other studies as indicated. In addition, at the end of follow-up, serum cystatin C, aldosterone and renin levels were measured. Also, 24-h urine production was monitored to determine creatinine, calcium and citrate levels, and proteinuria. GFR was calculated from serum creatinine by Schwartz’s formula [7] and from cystatin C by Filler’s formula [8]. Urine cytology and cystoscopy were performed in all patients at the end of follow-up. Two biopsies were obtained in the intestine close to the vesicointestinal anastomosis. All specimens were fixed in 10% formalin, embedded in paraffin and examined using haematoxylin and eosin staining.

Results Average age at time of enterocystoplasty was 11.8 years (range 3e18) and the mean follow-up was 11 years (range 8e14.5). A total of 22 augmentations were performed

Table 2

Indications for enterocystoplasty No. of patients

Unilateral Bilateral Total (patients) Total (ureters) VUR 11 UHN 4

6 3

17a 7a

23 10

Eight patients had a normal upper urinary tract. Of the 23 refluxing ureters, 18 (78%) were high grade. a Three patients had VUR in one kidney and UHN in the other kidney.

No urodynamic improvement No VUR and/or UHN improvement Recurrent UTIs Associated to AUS implantation Bladder compliance worsening after AUS placement

29 21 14 4 2

Long-term follow-up of enterocystoplasty with ileum, and seven with sigmoid colon. Mean patient age at the end of this study was 22.2 years (range 13.2e31). In the urodynamic evaluations, all augmented bladders showed significant improvement in bladder compliance. Mean bladder capacity after enterocystoplasty was 521 ml (range 300e1000) at a mean detrusor pressure of 10 cm of water (range 5e15). This bladder compliance improvement resolved the UHN in all patients and VUR in 13/17 patients (76.4%). In one patient VUR improved from bilateral high grade to bilateral low grade, but VUR grade did not change in the other three patients (uni- and bilateral low grade, bilateral high grade). The patient with a bilateral high-grade VUR was subsequently treated effectively with an endoscopic sub-trigonal injection. During the course of this long-term follow-up there has been no new scarring on DMSA scintigraphy. Three patients presented bladder stones that required surgical intervention (10.3%, one patient twice), one patient had a symptomatic UTI because of temporary non-compliance with CIC, one patient had an intestinal obstruction, and another patient with an AUS required a continent catheterisable channel made from ileum because of problems with CIC through the urethra. At the end of this study, only one patient (3.4%) had mild metabolic acidosis that required alkali therapy, but this patient already had metabolic acidosis before enterocystoplasty. RF according to serum creatinine level and its corresponding GFR were normal in all patients. The serum cystatin C level was normal in 27 patients (less than 0.9 mg/l) and was elevated in two (1.38 mg/l). The GFR estimated from the serum cystatin C level was also normal in all patients, except in the two patients with an elevated serum cystatin C level (GFR below 80 ml/min per 1.73 m2). Therefore, based on serum creatinine level, all patients had a normal RF but, based on serum cystatin C level, two patients had mild renal impairment. Significant proteinuria (more than 4 mg/m2 per hour) was identified in 20/25 (80%) patients and in some of them it spontaneously disappeared. Urinary calcium excretion was in the normal range in all patients except two (more than 4 mg/kg per day), while 17/24 (71%) had hypocitraturia (urine citrate excretion less than 300 mg/day per 1.73 m2). Serum fasting concentrations of renin and aldosterone were lower than the mean normal values in 14/17 patients (82%) (renin below 3 ng/ml per hour and aldosterone below 250 pg/ml). Urine cytology and cystoscopy were normal in all patients and no malignant lesions have been found in the biopsy specimens. Results of histological analysis of the specimens are shown in Table 3. Only one patient has had some trouble with CIC through the urethra during follow-up, and all patients are dry day and night. In three patients CIC is not necessary to empty their bladders adequately. Most patients (90%) have had asymptomatic bacteriuria and one patient has had mild hypertension.

Discussion Enterocystoplasty is a very effective tool to avoid upper urinary tract and RF deterioration in NB with poor compliance and an unsatisfactory response to CIC and anti-

29 Table 3

Histological findings CI VA MH CIþVA CIþMH CIþVAþMH ED EDþVA

Ileum 4 1 1 Colon 2 e e sigmoid

5 e

1 1

4 e

2 1 e e

Six patients were without relevant histological findings (two ileum, four colon sigmoid). In one patient (ileum) the biopsy material specimen was insufficient for histological study. CI, chronic inflammation; VA, villous atrophy; MH, mucosal hyperplasia; ED, epithelial dysplasia.

cholinergic therapy. In the last few years several clinical studies have suggested that aggressive early non-operative intervention may decrease the need for bladder augmentation [9]. Fifty-two percent of our patients with VUR and/ or UHN were finally referred to us from other centres, which is why the average age for starting with CIC and anti-cholinergic therapy was so late (mean age 4.4 years). It is possible that in some of these patients enterocystoplasty might have been avoided if this treatment had been started earlier. Some authors have demonstrated that an anti-reflux procedure need not be performed at the time of bladder augmentation in NB with VUR or UHN. The rationale is that these patients have acquired VUR or UHN as a consequence of high intravesical pressure, and the condition may be reversed by correcting this [10e12]. In our patients with VUR or UHN, we did not make any attempt to re-implant the ureters and, after enterocystoplasty, UHN disappeared in all and VUR in 76.4% (13/17 patients). Deterioration of RF after enterocystoplasty is very unusual and seems to be associated with a poor preoperative baseline RF. Using serum creatinine and Schwartz’s formula to calculate GFR has limitations, particularly in patients with spina bifida, because serum creatinine is dependent on body and muscle mass. Some authors have suggested that cystatin C is a better measure of RF in these patients than serum creatinine [13]. In our study some discrepancies were found when, at the end of follow-up, we compared serum creatinine with serum cystatin C levels. Based on serum cystatin C level, two patients had mild renal impairment, while according to serum creatinine all patients had normal RF. Consequently, more studies are necessary to determine the best parameter to measure kidney function in these patients. Our study revealed two findings in some patients that have not been mentioned in other similar studies and require discussion: the presence of significant proteinuria and the low fasting serum concentrations of renin and aldosterone. Although proteinuria was significant during follow-up in 80% of our patients, in some of them it spontaneously disappeared. This raises the question as to whether the proteinuria was the result of hyperfiltration injury or that these proteins were actually being excreted by the implanted bowel segment. If the former is true, a linear increase in proteinuria during follow-up could justify treating the patient with angiotensin-converting enzyme inhibitors to avoid RF deterioration. The fasting serum concentration of renin and aldosterone was low in 82% of our patients without other

30 electrolyte abnormalities. The aetiology is unknown but we could speculate that changes in intracranial pressure (all patients had a ventriculoperitoneal shunt) may have affected vasopressin secretion and could be implicated. This question merits future study. Metabolic disturbances in patients with enterocystoplasty are the result of an unphysiological exposure of intestinal segment to urine. There is some evidence that the absorptive properties of the intestinal segment may decrease over time. Certainly, histological changes do occur in the intestinal mucosa, including mucosal atrophy and decreases in villi height; these changes have been suspected as potential mechanisms for decreased intestinal absorptive capacity [14]. Eleven of our patients presented mild villous atrophy on biopsy. Metabolic acidosis is mainly caused by re-absorption of ammonium chloride and, to a much lesser extent, secretion of bicarbonate. Metabolic acidosis is a common problem in some patients after enterocystoplasty, but usually is not severe enough to require alkali therapy [14], and a GFR of 55 ml/min per 1.73 m2 is sufficient to process the increased acid load [15]. As in other studies [16], none of our patients have experienced metabolic acidosis, except the one patient who presented with this condition before enterocystoplasty. This low incidence of metabolic acidosis could be attributed to the use of small intestinal segments (15e20 cm), frequent emptyings of the neobladder (4e5 times a day) and a GFR above 55 ml/min per 1.73 m2. Spontaneous bladder perforation is one of the most serious complications that can occur after enterocystoplasty. It has been reported to occur in 6e13% of augmented patients, and seems to be independent of the intestinal segment used. The risk is decreased by the presence of a continent catheterisable channel, probably because this makes compliance with CIC easier [17]. Fortunately, there were no perforations in the present series. Bladder stones are a common problem after enterocystoplasty. It is generally accepted that the combination of urinary stasis, increased mucous production, non-compliance with CIC, abnormalities in urine composition and chronic infection with urea-splitting organisms causes the high incidence of bladder stones, reported as ranging between 10% and 52%. This wide range depends on how many gastrocystoplasties are included [18]. In series with a preponderance of ileocystoplasties the rate has been reported to be as high as 52% [19]. Enterocystoplasty management should attempt to prevent stone formation. Hensle et al. [20] demonstrated that the use of a standard prophylactic irrigation protocol significantly decreased the rate of calculous formation from 43% to 7% in enterocystoplasty patients. It is also important to mention that in that study the patients with an abdominal stoma had a significantly greater risk of stone formation than those using the native urethra (66% vs. 15%). This observation has also been made by other authors [21]. Asymptomatic bacteriuria in these patients is nearly universal (90% of our patients) and is generally of no clinical relevance and should not be treated with antibiotics unless urine culture indicates urea-splitting organisms. Urine citrate levels are frequently low in these patients (71% of our patients) and oral citrate should be taken to decrease the risk of stone formation. The low incidence of bladder stones found in our study (10.3%) could

P. Lo ´pez Pereira et al. be related to strict compliance with the daily irrigation protocol and the fact that only one patient had an abdominal stoma. The risk of malignancy in an augmented bladder is an increasing concern. The majority of reported cases have occurred in adults with multiple risk factors, making it difficult to determine the exact independent risk associated with enterocystoplasty. Aetiology is multifactorial involving chronic inflammation and carcinogens (nitrosamines and growth factors), and the bowel segment used does not seem to be relevant. Most reported tumours have been adenocarcinomas, but transitional cell carcinoma and others have also been reported [3,22]. Soergel et al. [3] estimate a prevalence of 1.2% (10-year lag time) in their database of 483 augmented patients, similar to the 2% prevalence reported by Shokeir et al. [23] in 186 augmented patients followed for more than 10 years. Some tumours have occurred within 5 years of augmentation [22], so an annual cystoscopy is recommended, beginning 5e10 years after enterocystoplasty, and the role of urine cytology needs to be determined for the future. Fortunately, up to the present time, cystoscopy in all of our patients has been normal and no malignant lesion has been found in the biopsy specimens.

Conclusion Enterocystoplasty is an excellent treatment option for NB refractory to conservative treatment. This procedure preserves RF and resolves the UHN and/or VUR in most patients without re-implanting the ureters. The incidence of some of the complications associated with this procedure can be decreased with a strict catheterisation regimen, daily bladder irrigation protocol and treatment of hypocitraturia. As the possible complications are not insignificant, including the potential risk of malignancy, these patients need close lifelong follow-up and annual cystoscopy from 5 to 10 years after augmentation. Further questions to be answered in the treatment of these patients are the significance of the proteinuria and low fasting serum concentrations of renin and aldosterone that some present, as well as the best parameter for RF evaluation.

References [1] Vajda P, Kaiser T, Magyarlaki A, Farkas AMV, Pinter AB. Histological findings after colocystoplasty and gastrocystoplasty. J Urol 2002;168:698e701. [2] Quek ML, Ginsberg DA. Long-term urodynamics followup of bladder augmentation for neurogenic bladder. J Urol 2003; 169:195e8. [3] Soergel TM, Cain MP, Misseri R, Gardner TA, Koch MO, Rink RC. Transitional cell carcinoma of the bladder following augmentation cystoplasty for the neuropathic bladder. J Urol 2004; 172:1649e52. [4] Hayashi Y, Yamataka A, Kaneyama K, Kato Y, Lane GJ, Miyano T. Review of 86 patients with myelodysplasia and neurogenic bladder who underwent sigmoidocolocystoplasty and were followed more than 10 years. J Urol 2006;176: 1806e9. [5] Bramble FJ. The treatment of adult enuresis and urge incontinence by enterocystoplasty. BJU Int 1982;54:693e6.

Long-term follow-up of enterocystoplasty [6] Mundy AR, Stephenson TP. ‘‘Clam’’ ileocystoplasty for the treatment of refractory urge incontinence. BJU Int 1985;57:641e6. [7] Schwartz GJ, Brion LP, Spitzer A. The use of plasma creatinine concentration for estimating glomerular filtration rate in infants, children and adolescents. Pediatr Clin North Am 1987; 34:571e90. [8] Filler G, Lepage N. Should the Schwartz formula for estimation of GFR be replace by cystatin C formula? Pediatr Nephrol 2003;18:981e5. [9] Kaefer M, Pabby A, Kelly M, Darbey M, Bauer SB. Improved bladder function after prophylactic treatment of the high risk neurogenic bladder in newborns with myelomeningocele. J. Urol 1999;162:1068e71. [10] Nasrallah PF, Alibadi HA. Bladder augmentation in patients with neurogenic bladder and vesicoureteral reflux. J. Urol 1991;146:563e6. [11] Lopez Pereira P, Martinez Urrutia MJ, Lobato Romera R, Jaureguizar E. Should we treat vesicoureteral reflux in patients who undergo bladder augmentation for neuropathic bladder? J. Urol 2001;165:2259e61. [12] Soylet Y, Emir H, Ilce Z, Yesildag E, Buyukunal SNC, Danismend N. Quo vadis? Ureteric reimplantation or ignoring reflux during augmentation cystoplasty. BJU Int 2004;94:379e80. [13] Damidharka VR, Kwon C, Stevens G. Serum cystatin C is superior to serum creatinin as a marker of kidney function; A metaanalysis. Am J Kidney Dis 2002;40:221e6. [14] Gilbert SM, Hensle TW. Metabolic consequences and long-term complications of enterocystoplasty in children: a review. J. Urol 2005;173:1080e6.

31 [15] Kristansson A, Davidsson T, Mansson W. Metabolic alterations at different levels of renal function following continent urinary diversion through colonic segments. J. Urol 1997;157: 2099e103. [16] Hafez AT, McLorie G, Gilday D, et al. Long-term evaluation of metabolic profile and bone mineral density after ileocystoplasty in children. J. Urol 2003;170:1639e42. [17] Metacalfe PD, Casale AJ, Kaefer MA, et al. Spontaneous bladder perforation: A report of 500 augmentations in children and analysis of risk. J. Urol 2006;175:1466e71. [18] DeFoor W, Minevich E, Reddy P, et al. Bladder calculi after augmentation cystoplasty: risk factors and prevention strategies. J. Urol 2004;172:1964e6. [19] Palmer LS, Franco I, Kogan SJ, Reda E, Gill B, Levitt SB. Urolithiasis in children following augmentation cystoplasty. J. Urol 1993;150:726e9. [20] Hensle TW, Bingham J, Lam J, Shabsigh A. Preventing reservoir calculi after augmentation cystoplasty and continent urinary diversion: the influence of an irrigation protocol. BJU Int 2004;93:585e7. [21] Metacalfe PD, Cain MP, Kaefer M, et al. What is the need for additional bladder surgery after bladder augmentation in childhood? J. Urol 2006;176:1801e5. [22] Austen M, Ka ¨lble T. Secondary malignancies in different forms of urinary diversion using isolated gut. J. Urol 2004; 172:831e8. [23] Shokeir AA, Shamaa M, El-Mekresh MM, El-Baz M, Ghoneim MA. Late malignancy in bowel segments exposed to urine without fecal stream. Urology 1995;45:657e61.