This is the first study that examines the outcome of augmentation cystoplasty in children with stages III and IV chronic kidney disease secondary to neurogenic bladder in whom transplantation was not imminent. 

Spina bifida, anorectal malformation and spinal cord injury are causes of neurogenic bladder that portend a high risk of renal damage. Goals of therapy include ensuring safe storage and emptying usually by a combination of antimuscarinics (or botulinum toxin injections) and intermittent catheterization (IC). However, some children especially those with poorly compliant bladders require augmentation cystoplasty. Good renal function has been considered important for handling the metabolic abnormalities that occurs due to reabsorption by the bowel segment. Unfortunately, we continue to see children with preventable secondary kidney damage due to a delay in institution of therapy. This poses a dilemma since poor bladder storage function could potentially accelerate kidney damage in these patients hastening the onset of end stage kidney disease. This is a retrospective study from a tertiary care center.

Hospital records of all children with neurogenic bladder who underwent augmentation cystoplasty between January 2008 and December 2017 were retrieved to identify those with chronic kidney disease stages III and IV at the time of surgery. Records were studied for outcome and complications including the number of febrile urinary infections requiring admission, kidney function (eGFR, National Kidney Foundation Calculator) and somatic growth (percentiles, compared with population data). Height-independent eGFR estimate was used for one girl with severe kyphoscoliosis. Ethics committee approval was taken. Descriptive statistics and Mann Whitney U test were calculated using R (ver 3.1.3).

As part of departmental policy, all children presenting with renal dysfunction undergo aggressive conservative management with initial indwelling catheter and subsequent urodynamics-based management with antimuscarinics and clean IC. Augmentation cystoplasty is offered as a last resort typically in those with unresponsive poorly compliant bladders. Regardless of any secondary vesicoureteral reflux, history of febrile urinary infections or sphincteric incompetence, children are not offered ureteric reimplantation or continence procedures at augmentation. All augmentations are carried out using de-tubularised ileum fashioned into a cup.

Augmentation cystoplasty was performed on 79 children with neurogenic bladder of which 13 (10 spinal dysraphism, 3 anorectal malformation) were in children with secondary chronic kidney disease stages III and IV (10 girls; range 3.0y to 14.9y; median 8.0y). All patients underwent ileocystoplasty and the median length of bowel used was 25cm (IQR 25-30cm). One had simultaneous nephrectomy and one a mitrofanoff conduit. Incontinence episodes were experienced by 10 of the 13 children despite maximum dose of antimuscarinics. Bilateral hydroureteronephrosis was noted in all. Vesicoureteral reflux was recorded in seven children. On urodynamics, median bladder capacity was 74% of expected capacity for age  (IQR 53%-89%) and median compliance was 5ml/cm H20 (IQR 2-6ml/cm H20). 

There was resolution of incontinence in all children. One girl who had a simultaneous appendiceal mitrofanoff (her urethral meatus was recessed) needed re-hospitalization at 2 weeks for dyselectrolytemia and acute on chronic kidney disease that resolved. One girl required cystolithotripsy at four years. One boy had urethral trauma at 11 months following IC with a false passage near the verumontanum that healed uneventfully following cystoscopy and a short period of indwelling catheterization. 

The eGFR was 24ml/min/1.73m2 (IQR 20ml/min/1.73m2- 44ml/min/1.73m2) at initial presentation to the department, which improved significantly to 52ml/min/1.73m2 (IQR 43ml/min/1.73m2- 55ml/min/1.73m2; p=0.004) with conservative management. This improved further to 61ml/min/1.73m2 (IQR 48.5ml/min/1.73m2- 72ml/min/m2) one year following augmentation cystoplasty (p=0.036). No further improvement was noted at any subsequent follow up with median eGFR 61, 57, 60, 58, 68 and 59ml/min/1.73m2 after 2-7 years following surgery respectively (Figure 1). Three children had an eGFR at last follow up that was numerically below the preoperative eGFR (-1, -4 and -12 ml/min/1.73m2). Four children were hypertensive at surgery and continued using the same dose of medication at last follow up.

Somatic growth impairment was noted in a majority of children at presentation for surgery with improvements in most children by the last follow up (Figure 2). Reduction in hydronephrosis was noted in all. There was a reduction in febrile urinary tract infections requiring admission from 0.72 per patient-year to 0.04 per patient-year following surgery.

Traditionally, poor renal function has been considered a relative contraindication for augmentation in children not facing imminent transplantation. This study shows that augmentation cystoplasty in select children is safe and leads to modest improvements in renal function and somatic growth. The procedure also serves to prepare the lower urinary tract for the inevitable renal transplantation that many of these children will eventually face. A standardized protocol and avoiding additional surgery might have helped us keep the complication rate relatively low in this high-risk group.

Renal function showed modest improvement following augmentation cystoplasty but this was limited to the first year. Most patients showed stunting of height and weight at surgery but there was improvement following augmentation. Despite this improvement, they remained behind their age-matched peers. 

None of the children were given antimuscarinics following surgery. Augmentation was effective at resolving urinary incontinence although we did not perform continence procedures in any of these children. It is possible that the high end-fill pressures in most of these patients (as evidenced by the very low recorded compliance) marked out a group of patients in whom the outlet resistance was high and hence, once storage pressures became low, continence was restored. 

We did not reimplant any of the ureters regardless of reflux. The sharp reduction in urinary tract infection episodes, resolution of hydronephrosis and improvement in renal function all attest to the fact that ureteric reimplantation may not be required in children with reflux secondary to neurogenic bladder. 

Delay in institution of appropriate therapy has been shown to be associated with subsequent poor bladder compliance, a risk factor for kidney damage. This study shows that such patients often benefit from aggressive urodynamics-based conservative management even when they present late. In fact, initial renal function more than doubled with conservative management alone.

Bladder capacity is not a reliable surrogate for safety in this patient population. Several of these children had normal bladder capacity but unsafe pressures. Typically, such children need shorter bowel segments of about 25cm for augmentation.

Limitations include lack of control group with reimplantation, no postoperative reflux evaluation and small numbers.

Augmentation cystoplasty can be carried out with acceptable morbidity in children with chronic kidney disease stages III and IV secondary to neurogenic bladder. Surgery is effective with improvements in renal function, somatic growth and propensity for febrile infections. Ureteric reimplantation might not be necessary.