The effect of intestinal urinary reservoirs on renal function: a 10-year follow-up
Objective To study the effect of the storage of urine in intestinal reservoirs on long-term renal function and the possible causes of deterioration.
Patients and methods Eighty-seven patients (aged 4–35 years) with bladder exstrophy who underwent reconstruction of the lower urinary tract using a bowel segment were enrolled in a prospective protocol. The glomerular filtration rate (GFR) was measured before and after surgery at 1, 2, 5 and 10 years using 51Cr-ethylenediamine tetra-acetic acid. Patients with a decline in GFR of > 5% were investigated to identify the cause.
Results Of 58 patients with a follow-up of ≥ 10 years, 53 were evaluable, four having been lost to follow-up and one refusing to accept the protocol. In these 53 patients, the mean ( sd) GFR decreased from 97.9 (20.4) to 92.9 (23.6) mL/min/1.73 m2 (P = 0.24). However, this decrease was accounted for by 10 patients (19%) whose GFR fell by ≥ 20% over the 10 years. The causes of renal deterioration in these 10 patients were; chronic retention and/or infection caused by inadequate catheterization in poorly compliant patients (five), uretero-ileal stenosis (one), a high-pressure reservoir (one) and uncertain causes (three).
Conclusions For 80% of the patients, the storage of urine in intestinal reservoirs did not change renal function for at least 10 years. However, ≈ 20% of patients had some deterioration in renal function during the 10-year follow-up, usually from identifiable and remediable causes. The storage of urine in bowel does not appear to be inherently damaging to kidney function. Patients with an enterocystoplasty need regular monitoring of renal function; when deterioration is detected the urinary tract must be functionally assessed.
Segments of bowel are increasingly used for bladder augmentation, replacement of the bladder in situ after cystectomy and continent urinary diversion. A variety of operative procedures has been described using either ileum or colon or a combination of both. These techniques have been proposed as alternatives to conduits, to avoid an external appliance.
The most important requirement for any bladder substitute is that it should not jeopardize the integrity of the upper urinary tract. Long-term results of urinary diversion, e.g. ureterosigmoidostomy and ileal conduits, showed some deterioration of renal function from several causes, including obstruction of urine flow, reflux, infection and stone disease [1,2]. In these patients, renal deterioration was usually caused by the complications of the diversion. In the absence of complications, the ileal conduit does not seem to be intrinsically harmful to the kidneys. Perhaps for this reason the technique of conduit construction has hardly changed in the 50 years since its original description. On the other hand, the high pressure in a ureterosigmoidostomy is probably responsible for the high rate of renal damage.
The question then arises of the safety of urine storage in detubularized intestinal reservoirs. There are a few studies of patients with intestinal storage reservoirs that report values for renal function during the follow-up [3–5]. In general, renal function is well preserved and compares favourably with that after conduit diversion.
To monitor the outcome of lower urinary tract reconstruction, a prospective study was started in 1983 at the authors' institution, and included all patients with any form of enterocystoplasty. As part of this study, renal function was monitored by measuring the GFR. This report includes only patients with exstrophy whose condition is generally an isolated one of urinary storage with no other problems that might alter renal function. At the time that the protocol was started, ethical approval was not required. The Ethical Committee of University College London Hospitals (of which the authors' institution is now a part) has given permission to publish this report.
Patients and methods
Since 1983, 87 patients with bladder exstrophy who underwent reconstruction of the lower urinary tract using detubularized bowel segments have been enrolled in a prospective study. Details of the protocol have been published elsewhere . In 58 patients the follow-up was ≥ 10 years. Four of the patients were lost to follow-up or had inadequate data, and one patient refused to participate in the protocol. Thus, the study group comprised 53 patients (35 male, 18 female) with a diagnosis of classical exstrophy and a follow-up of ≥ 10 years. Their mean (range) age at operation was 14.9 (4–35) years. Of the 53 patients, 28 had undergone conversion from an incontinent urinary diversion or ureterosigmoidostomy ( Table 1) and six had a single kidney at the construction of the reservoir; all kidneys were normal at birth.
Table 1. Type and duration of urinary diversion before cystoplasty
|Sigmoid conduit||12||14 (9–26)|
|Ileal conduit||11||19 (10–30)|
|No previous diversion||25|
Various methods of urinary reconstruction were used, i.e. augmentation or substitution cystoplasty using detubularized bowel segments (ileum, caecum, sigmoid). The outlet and continence mechanisms were the urethra and its sphincter, an appendix or ileal tube in a Mitrofanoff tunnel, Kock nipple or ileocaecal valve ( Table 2). Ureteric implantation was either into the bladder or into the bowel segment. Where the ureter was implanted into the intestinal segments, an antireflux technique was not systematically used and the type of implantation was sometimes unknown because of previous surgery.
Table 2. Procedures undertaken for the reservoir and continence mechanism
|Bladder + Sigmoid||Urethra||11|
|Bladder + Ileum||Mitrofanoff||2|
|Bladder + Ileum + Caecum||Urethra||1|
|Bladder + Ileum + Sigmoid||Mitrofanoff||2|
|Ileum + Sigmoid||Mitrofanoff||10|
|Ileum + Caecum||Mitrofanoff||1|
|Sigmoid + Caecum||Mitrofanoff||1|
All patients (or their parents) were informed before surgery of the intended follow-up protocol; this comprised, among other investigations, measurements of blood pressure, electrolytes, a full blood count, renal ultrasonography and plain abdominal films.
For renal function, serum creatinine levels were measured annually. The GFR (corrected to 1.73 m2 body surface area) was determined by plasma clearance of 51Cr-EDTA, using the single-shot method [7,8]. The GFR was measured before surgery and at 1, 2 and 5 years afterwards, and then at 5-year intervals. The measurement of GFR was repeated yearly if it was unstable. If renal function deteriorated the urinary tract was assessed functionally, including urine culture and protein estimation, IVU or 99mTc-MAG3 renal scintigraphy, videocystometrography (VCMG) and a nephrological review, as indicated.
In the 53 patients, the GFR decreased from a mean (median, sd) of 97.9 (100, 20.4) to 92.9 (97, 23.6) mL/min/1.73 m2 (P = 0.24). However, in 43 of the 53 patients (including two with an initial GFR of < 50 mL/min/1.73 m2) there was no significant deterioration. The decrease in GFR was almost wholly accounted for by 10 patients (19%) whose GFR after surgery decreased by ≥ 20% during the 10-year follow-up ( Table 3).
Table 3. Causes of deterioration of renal function in 10 patients with a decrease in GFR of ≥ 20%
|1||Ileum patch/Mitrofanoff||66||27||High pressure reservoir||No|
|2||Ileum + Sigmoid patch/Mitrofanoff||100||73||Unknown (+ single kidney?)||Sigmoid loop (1978–88)|
|3||Ileum + Sigmoid patch/Mitrofanoff||115||78||Stones + recurrent pyelonephritis||Sigmoid loop (1977–87)|
|4||Ileum + Sigmoid pouch/Mitrofanoff||120||86||Renal stone + UTIs||Sigmoid loop (1970–85)|
|5||Ileum pouch/Mitrofanoff||64||50||Uretero-ileal stricture|
(+ single kidney?)
Ileal loop (1971–89)
|6||Ileum + Caecum patch/Urethra||126||74||Unknown||No|
|7||Ileum pouch/Mitrofanoff||100||62||Renal stone + UTIs||Ileal loop (1967–86)|
|8||Sigmoid pouch/Mitrofanoff||89||72||Stones + recurrent pyelonephritis||Sigmoid loop (1974–89)|
|9||Ileum + Sigmoid pouch/Mitrofanoff||73||55||Hydronephrosis/|
|Ileal loop (1967–88)|
|10||Ileum pouch/Kock||115||82||Unknown||Ileal loop (1967–85)|
In four patients (nos 3, 4, 7 and 8) renal deterioration was caused by ureteric or renal stones from chronic infection with urea-splitting organisms. One patient (no. 1) had a grossly elevated reservoir pressure and one had an anastomotic ureteric stricture (no. 5); surgical correction stabilized renal function. Patient no. 9 had chronic hydroureter and hydronephrosis, with no evidence of reflux or obstruction and normal pouch dynamics. It is probable that his poor compliance with self-catheterization lead to chronic over-distension.
In three patients (nos. 2, 6 and 10) no certain cause was found; in one (no. 2) chronic renal infection was the probable cause as he underwent nephrectomy for a poorly functioning kidney, and histology showed chronic pyelonephritis. One patient failed to comply with the investigation programme. Ultimately, there was only one patient in whom no cause was established after full investigation for the significant deterioration in renal function.
The effects on renal function produced by the storage of urine in bowel have not been well documented. A few studies have addressed the issue [3–5,9,10]; in general, they show good preservation of renal function. However, most are retrospective, and some include few patients or a short follow-up. In some reports the patients are a heterogeneous group and there are factors other than the reconstruction (e.g. cancer or neurogenic bladder) that might affect long-term renal function.
The present series is a homogeneous group of patients within a prospective study. Patients with bladder exstrophy have several advantages in this respect. Congenital renal damage is rare in patients with exstrophy before surgical correction, so that renal dysplasia or intrauterine nephropathy can be eliminated as a cause of any subsequent deterioration in renal function . Unlike patients with neurogenic bladder or PUV, dysfunction of the lower urinary tract does not change in childhood. Moreover, patients with bladder exstrophy usually undergo reconstruction during childhood or adolescence, allowing a long-term follow-up.
Renal function was assessed using the plasma clearance of 51Cr-EDTA; this is an ideal agent to measure GFR, because there is no binding to plasma constituents and no reabsorption or secretion by renal tubules. This method correlates well with the clearance of inulin , and is more reliable and sensitive than serum creatinine level and creatinine clearance rate. Serum creatinine level only increases when the GFR is < 40 mL/min. Creatinine is reabsorbed by ileal segments and only creatinine clearance measured under diuretic conditions correlates well with true renal function . Conversely, EDTA absorption from the intestinal segment in contact with urine is absent or minimal .
In patients with damage from post-renal causes, which would include reflux and infection in patients with bowel reservoirs, it is known that tubular damage precedes glomerular damage. Kristjansson et al. showed that the urinary protein HC may be a suitable marker for detecting early renal impairment. The present study started in 1983 and did not include an assessment of renal tubular function.
In any patient renal deterioration may be caused by several possible post-renal factors; the most obvious are ureteric obstruction and upper tract stones, which accounted for four of the present 10 patients.
Chronic urinary infection is promoted by the interposition of intestinal segments in the urinary tract ; most patients with bowel reservoirs have some bacteriuria. In a previous report on patients in this protocol, it was shown that 30% had troublesome infections and a further 17% had occasional symptomatic infections . Although symptomatic infections accounted for four (or possibly five) of the present patients, mainly by causing stones, the incidence of renal damage is much lower than would be expected if it was caused by bacteriuria alone. Antibodies to Escherichia coli and to Proteus are associated with small kidneys, suggesting that frequent systemic infections may cause renal destruction . We manage these infections aggressively, including re-education in catheterizing and washout techniques, and the liberal use of long-term prophylactic antibiotics. This may limit the renal damage, although it is possible that even frequent infections do not cause renal damage in this setting.
High storage pressure is a recognized cause of renal damage in several congenital anomalies. Although measurements of intestinal reservoir pressure are difficult to interpret, a high-pressure, poorly compliant reservoir was probably a cause of the renal damage in one of the present patients. Re-augmentation stabilized both renal and reservoir function. In a further patient, persistent failure to comply with the self-catheterization regimen appears to have caused chronic upper tract dilatation and renal damage. However, such behaviour has been noted in other patients in this protocol, apparently without causing renal damage. In two patients no cause for renal deterioration was identified, although in one the investigation was not completed.
These data show that lower urinary tract reconstruction was associated with a significant deterioration in renal function in 19% of patients after 10 years. However, in at least 80% of those with deterioration, there was a mechanical or remedial cause. There was no evidence that the storage of urine in intestinal reservoirs per se caused renal damage. With such a high rate of renal damage, close monitoring of the GFR is essential. When there is a significant decrease, a careful search must be made for a correctable cause. Conversion of the reservoir to an incontinent, cutaneous diversion should not be required.
E. Fontaine, MD, Specialist Registrar; currently, Service d'Urologie, Hôpital Ambroise Paré, Boulogne, France.
R. Leaver, RGN, BSc(Hons), Clinical nurse specialist.
C.R.J. Woodhouse, MB, FRCS, FEBU, Reader in Adolescent Urology.
C.R.J. Woodhouse, Institute of Urology and Nephrology, 48, Riding House Street, London, W1P 7PN, UK.