- Email: [email protected]
Risk Factors for End Stage Renal Disease in Children With Posterior Urethral Valves
Risk Factors for End Stage Renal Disease in Children With Posterior Urethral Valves William DeFoor,* Curtis Clark, Elizabeth Jackson, Pramod Reddy, Eugene Minevich and Curtis Sheldon From the Division of Pediatric Urology, Cincinnati Children’s Hospital, Cincinnati, Ohio
Purpose: Obstructive uropathy secondary to posterior urethral valves is an important cause of end stage renal disease in children. Early diagnosis and intervention to decrease bladder pressure and stabilize the upper urinary tract are important to delay or prevent the progression of renal insufficiency. We analyzed the records of patients with posterior urethral valves to determine risk factors that might be predictive of ultimate renal failure. Materials and Methods: A retrospective cohort study was performed of children presenting to our institution with a diagnosis of posterior urethral valves from 1975 to 2005. Patient demographics, clinical background, laboratory and radiographic data, and renal outcomes were abstracted from the medical record. Potential risk factors were analyzed, such as high grade vesicoureteral reflux at diagnosis, nadir serum creatinine greater than 1.0 mg/dl, urinary tract infection and severe bladder dysfunction requiring clean intermittent catheterization. Risk factors were analyzed by univariate analysis with Fisher’s exact test. Those achieving significance were placed in a multivariate logistic regression model and an OR was generated. Results: A total of 142 patients were identified, of whom half presented in the neonatal period. Of the patients 119 had sufficient records for evaluation and mean followup was 7.2 years. A total of 15 patients progressed to end stage renal disease. The mean interval from diagnosis to end stage renal disease was 8.1 years. Of these patients 93% initially presented with vesicoureteral reflux and 87% ultimately required clean intermittent catheterization. Increased nadir creatinine was seen in 80% of cases. Multivariate analysis revealed that increased nadir creatinine and bladder dysfunction were independent risk factors for end stage renal disease (OR 71 and 8.9, respectively). Vesicoureteral reflux was also associated with an increased risk of end stage renal disease (OR 2.0), although this was not statistically significant. Urinary tract infections were not associated with end stage renal disease. Conclusions: Patients with posterior urethral valves and severe bladder dysfunction in whom nadir creatinine remains increased are at risk for upper urinary tract deterioration, requiring renal replacement therapy. It is unclear whether high grade vesicoureteral reflux at diagnosis may also be a poor prognostic sign. Further analysis is necessary to evaluate the effects of early aggressive bladder management on renal outcomes. Key Words: posterior urethral valve, end stage renal disease, pediatric renal transplantation, risk factors
bstructive uropathy secondary to PUV is still an important cause of ESRD in children. According to the North American Pediatric Renal Trials and Collaborative Studies registry urological diseases, eg obstructive uropathy and renal aplasia/dysplasia, cause more than 30% of all chronic kidney disease and ESRD in children. Thus, they are the most common etiology of ESRD in this age group.1 Prompt diagnosis and bladder decompression are important early tenets in the treatment of these patients to stabilize the upper urinary tract. However, ongoing evaluation and intervention are necessary to delay or prevent progression to chronic renal insufficiency. We analyzed a cohort of patients with PUV to determine risk factors that might be predictive of the need for renal replacement therapy.
O
* Correspondence and requests for reprints: Division of Pediatric Urology, Cincinnati Children’s Hospital Medical Center, 3333 Burnett Ave., MLC 5037, Cincinnati, Ohio 45229-3039 (telephone: 513636-6758; FAX: 513-636-6753; e-mail: [email protected]).
0022-5347/08/1804-1705/0 THE JOURNAL OF UROLOGY® Copyright © 2008 by AMERICAN UROLOGICAL ASSOCIATION
MATERIALS AND METHODS A retrospective cohort study was performed of all children presenting to a single pediatric institution with a clinical and radiographic diagnosis of a posterior urethral valve from 1975 to 2005. Inclusion criteria were an initial renal and bladder ultrasound as well as a fluoroscopic voiding cystourethrogram that confirmed the diagnosis. Patient demographics, clinical background, laboratory and radiographic data, and renal functional outcomes were abstracted from the medical record and tabulated into a database. Prenatal diagnosis was recorded when available. The progression to renal failure was confirmed by the division of pediatric nephrology ESRD database at our institution. Clinical characteristics thought to be important in the progression to ESRD were chosen by clinical judgment and previously published reports.2,3 The risk factors chosen for this analysis were high grade VUR at diagnosis (grade 3 or higher), nadir serum creatinine greater than 1.0 mg/dl after a period of bladder decompression as a proxy for congenital renal dysplasia, recurrent febrile urinary tract infections
1705
Vol. 180, 1705-1708, October 2008 Printed in U.S.A. DOI:10.1016/j.juro.2008.03.090
1706
RISK FACTORS FOR END STAGE RENAL DISEASE AND POSTERIOR URETHRAL VALVES
(greater than 2 symptomatic episodes of symptomatic bacteriuria during followup) and severe bladder dysfunction requiring CIC. An assessment of hydroureteronephrosis was not chosen for analysis because this was uniformly present in all patients at diagnosis and, thus, it was not helpful for discriminating adverse outcomes among the study group. The diagnosis of ESRD was recorded when the patient ultimately required dialysis or underwent renal transplantation. Patients undergoing living related donor transplantation before the absolute need for peritoneal or hemodialysis were included in the ESRD group. Data were recorded in binary form and compared between those with and without ESRD by univariate analysis with Fisher’s exact test. The probability of excluding the null hypothesis by chance (no difference between the 2 groups) was considered at p ⬍0.05. Risk factors achieving statistical significance were then analyzed in a multivariate logistic regression model using StatView®. The OR, CI and p value were generated. Risk factors that remained statistically significant while controlling for the other variables were believed to be independent predictors of ESRD.
RESULTS A total of 142 boys with PUV were identified during the study period at our institution, of whom 119 had sufficient records for evaluation to be included in the statistical analysis. The 23 remaining patients were lost to followup or relocated elsewhere and the missing data precluded a thorough analysis of the clinical course. Half of the patients presented in the neonatal period with a prenatally detected renal anomaly. The majority of the remaining patients, mostly those in the earlier half of the study period, presented with a febrile urinary tract infection. Mean followup was 7.2 years (range 3 to 24). Table 1 lists demographic data on the study group. Initial treatment in almost all cases was early bladder drainage and subsequent cystoscopy with transurethral incision of the obstructing valve leaflets after they were medically stabilized. Table 2 shows univariate analysis of the risk factors for ESRD in the 2 groups. A total of 15 patients (13%) in the cohort progressed to ESRD. Median age at progression to ESRD was 8.2 years (range 7 days to 17.5 years). Of these patients 93% initially presented with high grade VUR compared to only 48% of those who did not progress to ESRD (p ⬍0.001). VUR management was individualized but for the most part it was conservative and did not lead to early antireflux surgery. Upper tract urinary diversion, eg pyelostomy, was performed in only 6% of the overall cohort but it was more often necessary in those with worse pathological
TABLE 2. Univariate analysis of ESRD risk by risk factor No. ESRD (%) Yes Overall VUR Bladder dysfunction Urinary tract infection Nadir creatinine greater than 1.0 mg/dl
No
15 104 14 (93) 50 (48) 13 (87) 24 (23) 10 (67) 48 (46) 12 (80) 3 (3)
p Value (Fisher’s exact test) 0.0007 0.0001 0.17 ⬍0.0001
findings, ie 5 of 15 patients in the ESRD group and 2 of 104 in the nonESRD group. Of the 119 patients 37 ultimately were advised to begin CIC due to hostile bladder dynamics, which were leading to further renal injury such as progressive renal scarring or worsening hydronephrosis. Typical urodynamic findings in younger patients were a low capacity, poorly compliant bladder with high filling pressure. Radiographic abnormalities included a thick wall on ultrasound with severe bladder wall deformities such as trabeculations and diverticula but no persistent urethral obstruction. These younger patients requiring CIC were placed on anticholinergic medication if they were documented to have high bladder pressure on slow fill cystometrogram, defined as detrusor leak point pressure more than 40 cm H2O. Older patients who experienced detrusor myogenic failure and retentive bladder pathology generally did not require adjuvant bladder pressure lowering medication. In general patients were placed on antimicrobial chemoprophylaxis if they were found to have VUR, bladder dysfunction and/or persistent severe hydronephrosis. Of the patients 49% were treated for breakthrough recurrent urinary tract infections. This was not significant when stratified between the 2 groups (p ⬎0.1). Increased nadir creatinine greater than 1.0 mg/dl was seen in 13% of cases and it was usually associated with ultrasound characteristics of renal dysplasia. This was much more likely in the ESRD group (p ⬍0.0001). The 3 patients with persistently increased serum creatinine who did not yet require renal replacement therapy were followed closely for symptoms of chronic renal insufficiency. The remaining 101 patients had normal renal function. Multivariate logistic regression analysis revealed that increased nadir creatinine (OR 71, CI 10 – 482) and bladder dysfunction (OR 8.9, CI 1.1–73)) were independent risk factors for ESRD (table 3). VUR was also weakly associated with an increased risk of ESRD (OR 2.0, CI 0.2–24). However, since the lower CI for VUR was less than 1, the data could also be interpreted as reflux having a protective effect against the progression of renal failure. Recurrent urinary tract infections were not associated with an increased risk of chronic renal failure in this cohort.
TABLE 1. Patient demographics No. Pts Overall Neonatal diagnosis Followup (yrs) ESRD VUR Severe bladder dysfunction Urinary tract infection
119 60 7.2 15 64 37 58
TABLE 3. Multivariate logistic regression analysis of ESRD risk factors in patients with posterior urethral valves Risk Factor
OR* (95% CI)*
p Value
VUR Bladder dysfunction Nadir creatinine greater than 1.0 mg/dl
2.0 (0.2–24) 8.9 (1.1–73) 71 (10–482)
0.57 0.04 ⬍0.0001
* Controlling for other risk factors (1 equals no increased risk).
RISK FACTORS FOR END STAGE RENAL DISEASE AND POSTERIOR URETHRAL VALVES DISCUSSION Obstructive uropathy secondary to PUV is still a common cause of chronic renal insufficiency in children. The rate of progression to ESRD in these children has been documented to be 13% to 44% with more recent series on the lower end of the range.4 – 6 Our experience is similar to that in most modern series with 13% of our patients needing renal replacement therapy and with chronic renal insufficiency in several others. Nadir serum creatinine after a period of decompression has been helpful from a prognostic standpoint in several series. Drozdz et al found that patients with serum creatinine greater than 1.2 mg/dl before age 12 months progressed more rapidly to ESRD than those attaining this level later,2 while Bajpai et al reported similar findings using a cutoff of 0.8 mg/dl.7 Our study corroborated these findings but used a nadir creatinine threshold of 1.0 mg/dl not correlated with age. It has been well described that increased bladder pressure and external sphincter dyssynergia may have an adverse impact on upper tract integrity in patients with neuropathic voiding dysfunction as well as in patients with PUV. Upper tract deterioration and secondary VUR has correlated with intravesical pressures exceeding 40 cm H2O in children with myelomeningocele.8 Despite successful transurethral ablation patients with PUV may have persistent severe hydroureteronephrosis and stasis due to hostile bladder dynamics and inefficient bladder emptying. It is unclear whether this is acquired or caused by the initial bladder injury. Other investigators have reported the effect of bladder function on renal deterioration despite successful valve ablation. Ghanem et al identified poor compliance and detrusor overactivity as having a significant correlation with renal functional impairment, in addition to bilateral VUR and renal dysplasia.3 Bilateral VUR remained an independent risk factor in their series for a poor prognosis. Of note, only 20% of patients were deemed to have normal urodynamics. When poor drainage and urinary stasis persist after valve ablation, management options include high upper urinary tract diversion and aggressive intervention at the bladder level with CIC and anticholinergic medication. The role of urinary diversion is controversial but it may be necessary in ill and unstable patients with renal failure. In our series few patients underwent high urinary diversion unless they showed severe hydronephrosis, poor upper tract drainage and persistently increased serum creatinine. In contrast, a catheterization program was most consistently used in those with the most severe spectrum of disease. CIC can be helpful in patients with more chronic and slow renal deterioration due to increased bladder pressure. Additional drainage with nocturnal bladder emptying has been used to further decrease the bladder over distention associated with the polyuria of renal insufficiency.9 The patients in this cohort with poor renal outcomes tended to ultimately need treatment with CIC as well as anticholinergic medication. To our knowledge the effect of aggressive treatment at the bladder level on the rate of progression to ESRD has not been rigorously studied in a clinical trial. Renal transplan-
1707
tation into valve bladders has been shown to be safe in previous series.10 The limitations of this study include the retrospective nature of data acquisition, which may lead to missing data and possible selection bias. There may also be other known and unknown risk factors that were not included in the analysis that are important clinical factors for determining the renal prognosis. The cutoff for nadir serum creatinine at 1.0 mg/dl could also be construed as arbitrary but it seemed to discriminate the groups fairly well in this series. In addition, using serum creatinine instead of the glomerular filtration rate as a proxy for congenital renal dysplasia could be less precise but it was more uniformly available, particularly in the newborn period after valve ablation. Differentiating VUR into bilateral and unilateral presentations may have been important to document. Unilateral VUR with dysplasia has long been thought to be renoprotective of the contralateral nonrefluxing kidney, although this has recently become controversial.11,12 CONCLUSIONS Patients with PUV and severe bladder dysfunction in whom nadir creatinine remains increased seem at risk for further deterioration of the upper urinary tract and progression to ESRD. It is unclear whether high grade VUR at diagnosis may also be a poor prognostic sign. Further analysis is necessary to evaluate the effects of early aggressive bladder management with CIC and anticholinergic medication on renal outcomes.
Abbreviations and Acronyms CIC ESRD PUV VUR
⫽ ⫽ ⫽ ⫽
clean intermittent catheterization end stage renal disease posterior urethral valves vesicoureteral reflux
REFERENCES 1.
Seikaly M, Ho PL, Emmett L and Tejani A: The 12th Annual Report of the North American Pediatric Renal Transplant Cooperative Study: renal transplantation from 1987 through 1998. Pediatr Transplant 2001; 5: 215. 2. Drozdz D, Drozdz M, Gretz N, Mohring K, Mehls O and Scharer K: Progression to end-stage renal disease in children with posterior urethral valves. Pediatr Nephrol 1998; 12: 630. 3. Ghanem MA, Wolffenbuttel KP, De Vylder A and Nijman RJ: Long-term bladder dysfunction and renal function in boys with posterior urethral valves based on urodynamic findings. J Urol 2004; 171: 2409. 4. Smith GH, Canning DA, Schulman SL, Snyder HM 3rd and Duckett JW: The long-term outcome of posterior urethral valves treated with primary valve ablation and observation. J Urol 1996; 155: 1730. 5. Tejani A, Butt K, Glassberg K, Price A and Gurumurthy K: Predictors of eventual end stage renal disease in children with posterior urethral valves. J Urol 1986; 136: 857. 6. Ylinen E, Ala-Houhala M and Wikstrom S: Prognostic factors of posterior urethral valves and the role of antenatal detection. Pediatr Nephrol 2004; 19: 874. 7. Bajpai M, Dave S and Gupta DK: Factors affecting outcome in the management of posterior urethral valves. Pediatr Surg Int 2001; 17: 11.
1708
RISK FACTORS FOR END STAGE RENAL DISEASE AND POSTERIOR URETHRAL VALVES
8.
McGuire EJ, Woodside JR and Borden TA: Upper urinary tract deterioration in patients with myelodysplasia and detrusor hypertonia: a followup study. J Urol 1983; 129: 823. 9. Koff SA, Mutabagani KH and Jayanthi VR: The valve bladder syndrome: pathophysiology and treatment with nocturnal bladder emptying. J Urol 2002; 167: 291. 10. DeFoor W, Tackett L, Minevich E, McEnery P, Kitchens D, Reeves D et al: Successful renal transplantation in chil-
11.
12.
dren with posterior urethral valves. J Urol 2003; 170: 2402. Cuckow PM, Dinneen MD, Risdon RA, Ransley PG and Duffy PG: Long-term renal function in the posterior urethral valves, unilateral reflux and renal dysplasia syndrome. J Urol 1997; 158: 1004. Narasimhan KL, Mahajan JK, Kaur B, Mittal BR and Bhattacharya A: The vesicoureteral reflux dysplasia syndrome in patients with posterior urethral valves. J Urol 2005; 174: 1433.
DISCUSSION Dr. Gordon McLorie. In 1991 we published an article entitled “Nadir creatinine in posterior urethral valves.” It is either depressing or encouraging to note that the data still stand up and we have not come much further to predicting which children are going to end up with bad kidneys. Can you briefly answer what your major indications were for vesical or supravesical diversion in children with posterior urethral valves? Dr. William DeFoor. It is a difficult clinical decision. Generally, after valve ablation the serum creatinine remains high, or there is significant stasis or progressive hydronephrosis. I think in those children alone we would consider upper tract diversion but not as a primary treatment modality. Dr. Howard Snyder. I feel compelled to say one thing because a big Texan up in the sky whispered something in my ear. We fought this battle of supravesical diversion 20 years ago, and I thought it was dead. In his last paper before he died John Duckett made a plea to avoid supravesical diversion. We have had an abundance of experience with this and while vesicostomy lets a bladder continue to cycle and does not hurt it, supravesical diversion should never be done. There is no indication that a single patient has ever benefited by that over vesicostomy. I say that strongly because I think that is really the truth. There are too many young people here and I do not think they should get any mixed messages about this. Supravesical diversion prevents the bladder from cycling and if you want to cause harm to a patient with valves, do a supravesical diversion. Dr. Kenneth Glassberg. Gordon McLorie was talking about an important paper from Toronto on nadir creatinine, and other institutions have reported nadir creatinine as well. There is an important report by Kevin Burbidge who talked about continence at the age of 5. It is almost the same type of outcome because when you are continent at the age of 5 creatinine is much less. So continence and nadir creatinine seem to be important but can we affect either? I think all of these patients need urodynamics from the onset because we need to make sure that the valve is completely ablated. We should be doing videourodynamics and not VCUG. If we could affect continence by the age of 5 we will prevent renal failure in some patients or slow it down by age 10. We must be proactive. Right after valve ablation start thinking about anticholinergics because if you lower the pressures early in life, the natural tendency to develop myogenic failure does not occur. Also if you lower the pressure early, you might affect when or if end stage renal disease develops. Don’t just accept that dysplastic kidneys are going to result in end stage renal disease. We can have a positive affect. Dr. Joao Pippi Salle. I teach my fellows that occasionally there is a case for which ureterostomy or high diversion should be done based on a previous report by Jarrod Wisser published a few years ago. High diversion delayed the onset of end stage renal disease. There was no difference in bladder function and all patients had normal urodynamic parameters. Although I agree that valve ablation should be performed in most cases, high diversion should be considered in select cases. This is important for the young people to know.
Purpose: Obstructive uropathy secondary to posterior urethral valves is an important cause of end stage renal disease in children. Early diagnosis and intervention to decrease bladder pressure and stabilize the upper urinary tract are important to delay or prevent the progression of renal insufficiency. We analyzed the records of patients with posterior urethral valves to determine risk factors that might be predictive of ultimate renal failure. Materials and Methods: A retrospective cohort study was performed of children presenting to our institution with a diagnosis of posterior urethral valves from 1975 to 2005. Patient demographics, clinical background, laboratory and radiographic data, and renal outcomes were abstracted from the medical record. Potential risk factors were analyzed, such as high grade vesicoureteral reflux at diagnosis, nadir serum creatinine greater than 1.0 mg/dl, urinary tract infection and severe bladder dysfunction requiring clean intermittent catheterization. Risk factors were analyzed by univariate analysis with Fisher’s exact test. Those achieving significance were placed in a multivariate logistic regression model and an OR was generated. Results: A total of 142 patients were identified, of whom half presented in the neonatal period. Of the patients 119 had sufficient records for evaluation and mean followup was 7.2 years. A total of 15 patients progressed to end stage renal disease. The mean interval from diagnosis to end stage renal disease was 8.1 years. Of these patients 93% initially presented with vesicoureteral reflux and 87% ultimately required clean intermittent catheterization. Increased nadir creatinine was seen in 80% of cases. Multivariate analysis revealed that increased nadir creatinine and bladder dysfunction were independent risk factors for end stage renal disease (OR 71 and 8.9, respectively). Vesicoureteral reflux was also associated with an increased risk of end stage renal disease (OR 2.0), although this was not statistically significant. Urinary tract infections were not associated with end stage renal disease. Conclusions: Patients with posterior urethral valves and severe bladder dysfunction in whom nadir creatinine remains increased are at risk for upper urinary tract deterioration, requiring renal replacement therapy. It is unclear whether high grade vesicoureteral reflux at diagnosis may also be a poor prognostic sign. Further analysis is necessary to evaluate the effects of early aggressive bladder management on renal outcomes. Key Words: posterior urethral valve, end stage renal disease, pediatric renal transplantation, risk factors
bstructive uropathy secondary to PUV is still an important cause of ESRD in children. According to the North American Pediatric Renal Trials and Collaborative Studies registry urological diseases, eg obstructive uropathy and renal aplasia/dysplasia, cause more than 30% of all chronic kidney disease and ESRD in children. Thus, they are the most common etiology of ESRD in this age group.1 Prompt diagnosis and bladder decompression are important early tenets in the treatment of these patients to stabilize the upper urinary tract. However, ongoing evaluation and intervention are necessary to delay or prevent progression to chronic renal insufficiency. We analyzed a cohort of patients with PUV to determine risk factors that might be predictive of the need for renal replacement therapy.
O
* Correspondence and requests for reprints: Division of Pediatric Urology, Cincinnati Children’s Hospital Medical Center, 3333 Burnett Ave., MLC 5037, Cincinnati, Ohio 45229-3039 (telephone: 513636-6758; FAX: 513-636-6753; e-mail: [email protected]).
0022-5347/08/1804-1705/0 THE JOURNAL OF UROLOGY® Copyright © 2008 by AMERICAN UROLOGICAL ASSOCIATION
MATERIALS AND METHODS A retrospective cohort study was performed of all children presenting to a single pediatric institution with a clinical and radiographic diagnosis of a posterior urethral valve from 1975 to 2005. Inclusion criteria were an initial renal and bladder ultrasound as well as a fluoroscopic voiding cystourethrogram that confirmed the diagnosis. Patient demographics, clinical background, laboratory and radiographic data, and renal functional outcomes were abstracted from the medical record and tabulated into a database. Prenatal diagnosis was recorded when available. The progression to renal failure was confirmed by the division of pediatric nephrology ESRD database at our institution. Clinical characteristics thought to be important in the progression to ESRD were chosen by clinical judgment and previously published reports.2,3 The risk factors chosen for this analysis were high grade VUR at diagnosis (grade 3 or higher), nadir serum creatinine greater than 1.0 mg/dl after a period of bladder decompression as a proxy for congenital renal dysplasia, recurrent febrile urinary tract infections
1705
Vol. 180, 1705-1708, October 2008 Printed in U.S.A. DOI:10.1016/j.juro.2008.03.090
1706
RISK FACTORS FOR END STAGE RENAL DISEASE AND POSTERIOR URETHRAL VALVES
(greater than 2 symptomatic episodes of symptomatic bacteriuria during followup) and severe bladder dysfunction requiring CIC. An assessment of hydroureteronephrosis was not chosen for analysis because this was uniformly present in all patients at diagnosis and, thus, it was not helpful for discriminating adverse outcomes among the study group. The diagnosis of ESRD was recorded when the patient ultimately required dialysis or underwent renal transplantation. Patients undergoing living related donor transplantation before the absolute need for peritoneal or hemodialysis were included in the ESRD group. Data were recorded in binary form and compared between those with and without ESRD by univariate analysis with Fisher’s exact test. The probability of excluding the null hypothesis by chance (no difference between the 2 groups) was considered at p ⬍0.05. Risk factors achieving statistical significance were then analyzed in a multivariate logistic regression model using StatView®. The OR, CI and p value were generated. Risk factors that remained statistically significant while controlling for the other variables were believed to be independent predictors of ESRD.
RESULTS A total of 142 boys with PUV were identified during the study period at our institution, of whom 119 had sufficient records for evaluation to be included in the statistical analysis. The 23 remaining patients were lost to followup or relocated elsewhere and the missing data precluded a thorough analysis of the clinical course. Half of the patients presented in the neonatal period with a prenatally detected renal anomaly. The majority of the remaining patients, mostly those in the earlier half of the study period, presented with a febrile urinary tract infection. Mean followup was 7.2 years (range 3 to 24). Table 1 lists demographic data on the study group. Initial treatment in almost all cases was early bladder drainage and subsequent cystoscopy with transurethral incision of the obstructing valve leaflets after they were medically stabilized. Table 2 shows univariate analysis of the risk factors for ESRD in the 2 groups. A total of 15 patients (13%) in the cohort progressed to ESRD. Median age at progression to ESRD was 8.2 years (range 7 days to 17.5 years). Of these patients 93% initially presented with high grade VUR compared to only 48% of those who did not progress to ESRD (p ⬍0.001). VUR management was individualized but for the most part it was conservative and did not lead to early antireflux surgery. Upper tract urinary diversion, eg pyelostomy, was performed in only 6% of the overall cohort but it was more often necessary in those with worse pathological
TABLE 2. Univariate analysis of ESRD risk by risk factor No. ESRD (%) Yes Overall VUR Bladder dysfunction Urinary tract infection Nadir creatinine greater than 1.0 mg/dl
No
15 104 14 (93) 50 (48) 13 (87) 24 (23) 10 (67) 48 (46) 12 (80) 3 (3)
p Value (Fisher’s exact test) 0.0007 0.0001 0.17 ⬍0.0001
findings, ie 5 of 15 patients in the ESRD group and 2 of 104 in the nonESRD group. Of the 119 patients 37 ultimately were advised to begin CIC due to hostile bladder dynamics, which were leading to further renal injury such as progressive renal scarring or worsening hydronephrosis. Typical urodynamic findings in younger patients were a low capacity, poorly compliant bladder with high filling pressure. Radiographic abnormalities included a thick wall on ultrasound with severe bladder wall deformities such as trabeculations and diverticula but no persistent urethral obstruction. These younger patients requiring CIC were placed on anticholinergic medication if they were documented to have high bladder pressure on slow fill cystometrogram, defined as detrusor leak point pressure more than 40 cm H2O. Older patients who experienced detrusor myogenic failure and retentive bladder pathology generally did not require adjuvant bladder pressure lowering medication. In general patients were placed on antimicrobial chemoprophylaxis if they were found to have VUR, bladder dysfunction and/or persistent severe hydronephrosis. Of the patients 49% were treated for breakthrough recurrent urinary tract infections. This was not significant when stratified between the 2 groups (p ⬎0.1). Increased nadir creatinine greater than 1.0 mg/dl was seen in 13% of cases and it was usually associated with ultrasound characteristics of renal dysplasia. This was much more likely in the ESRD group (p ⬍0.0001). The 3 patients with persistently increased serum creatinine who did not yet require renal replacement therapy were followed closely for symptoms of chronic renal insufficiency. The remaining 101 patients had normal renal function. Multivariate logistic regression analysis revealed that increased nadir creatinine (OR 71, CI 10 – 482) and bladder dysfunction (OR 8.9, CI 1.1–73)) were independent risk factors for ESRD (table 3). VUR was also weakly associated with an increased risk of ESRD (OR 2.0, CI 0.2–24). However, since the lower CI for VUR was less than 1, the data could also be interpreted as reflux having a protective effect against the progression of renal failure. Recurrent urinary tract infections were not associated with an increased risk of chronic renal failure in this cohort.
TABLE 1. Patient demographics No. Pts Overall Neonatal diagnosis Followup (yrs) ESRD VUR Severe bladder dysfunction Urinary tract infection
119 60 7.2 15 64 37 58
TABLE 3. Multivariate logistic regression analysis of ESRD risk factors in patients with posterior urethral valves Risk Factor
OR* (95% CI)*
p Value
VUR Bladder dysfunction Nadir creatinine greater than 1.0 mg/dl
2.0 (0.2–24) 8.9 (1.1–73) 71 (10–482)
0.57 0.04 ⬍0.0001
* Controlling for other risk factors (1 equals no increased risk).
RISK FACTORS FOR END STAGE RENAL DISEASE AND POSTERIOR URETHRAL VALVES DISCUSSION Obstructive uropathy secondary to PUV is still a common cause of chronic renal insufficiency in children. The rate of progression to ESRD in these children has been documented to be 13% to 44% with more recent series on the lower end of the range.4 – 6 Our experience is similar to that in most modern series with 13% of our patients needing renal replacement therapy and with chronic renal insufficiency in several others. Nadir serum creatinine after a period of decompression has been helpful from a prognostic standpoint in several series. Drozdz et al found that patients with serum creatinine greater than 1.2 mg/dl before age 12 months progressed more rapidly to ESRD than those attaining this level later,2 while Bajpai et al reported similar findings using a cutoff of 0.8 mg/dl.7 Our study corroborated these findings but used a nadir creatinine threshold of 1.0 mg/dl not correlated with age. It has been well described that increased bladder pressure and external sphincter dyssynergia may have an adverse impact on upper tract integrity in patients with neuropathic voiding dysfunction as well as in patients with PUV. Upper tract deterioration and secondary VUR has correlated with intravesical pressures exceeding 40 cm H2O in children with myelomeningocele.8 Despite successful transurethral ablation patients with PUV may have persistent severe hydroureteronephrosis and stasis due to hostile bladder dynamics and inefficient bladder emptying. It is unclear whether this is acquired or caused by the initial bladder injury. Other investigators have reported the effect of bladder function on renal deterioration despite successful valve ablation. Ghanem et al identified poor compliance and detrusor overactivity as having a significant correlation with renal functional impairment, in addition to bilateral VUR and renal dysplasia.3 Bilateral VUR remained an independent risk factor in their series for a poor prognosis. Of note, only 20% of patients were deemed to have normal urodynamics. When poor drainage and urinary stasis persist after valve ablation, management options include high upper urinary tract diversion and aggressive intervention at the bladder level with CIC and anticholinergic medication. The role of urinary diversion is controversial but it may be necessary in ill and unstable patients with renal failure. In our series few patients underwent high urinary diversion unless they showed severe hydronephrosis, poor upper tract drainage and persistently increased serum creatinine. In contrast, a catheterization program was most consistently used in those with the most severe spectrum of disease. CIC can be helpful in patients with more chronic and slow renal deterioration due to increased bladder pressure. Additional drainage with nocturnal bladder emptying has been used to further decrease the bladder over distention associated with the polyuria of renal insufficiency.9 The patients in this cohort with poor renal outcomes tended to ultimately need treatment with CIC as well as anticholinergic medication. To our knowledge the effect of aggressive treatment at the bladder level on the rate of progression to ESRD has not been rigorously studied in a clinical trial. Renal transplan-
1707
tation into valve bladders has been shown to be safe in previous series.10 The limitations of this study include the retrospective nature of data acquisition, which may lead to missing data and possible selection bias. There may also be other known and unknown risk factors that were not included in the analysis that are important clinical factors for determining the renal prognosis. The cutoff for nadir serum creatinine at 1.0 mg/dl could also be construed as arbitrary but it seemed to discriminate the groups fairly well in this series. In addition, using serum creatinine instead of the glomerular filtration rate as a proxy for congenital renal dysplasia could be less precise but it was more uniformly available, particularly in the newborn period after valve ablation. Differentiating VUR into bilateral and unilateral presentations may have been important to document. Unilateral VUR with dysplasia has long been thought to be renoprotective of the contralateral nonrefluxing kidney, although this has recently become controversial.11,12 CONCLUSIONS Patients with PUV and severe bladder dysfunction in whom nadir creatinine remains increased seem at risk for further deterioration of the upper urinary tract and progression to ESRD. It is unclear whether high grade VUR at diagnosis may also be a poor prognostic sign. Further analysis is necessary to evaluate the effects of early aggressive bladder management with CIC and anticholinergic medication on renal outcomes.
Abbreviations and Acronyms CIC ESRD PUV VUR
⫽ ⫽ ⫽ ⫽
clean intermittent catheterization end stage renal disease posterior urethral valves vesicoureteral reflux
REFERENCES 1.
Seikaly M, Ho PL, Emmett L and Tejani A: The 12th Annual Report of the North American Pediatric Renal Transplant Cooperative Study: renal transplantation from 1987 through 1998. Pediatr Transplant 2001; 5: 215. 2. Drozdz D, Drozdz M, Gretz N, Mohring K, Mehls O and Scharer K: Progression to end-stage renal disease in children with posterior urethral valves. Pediatr Nephrol 1998; 12: 630. 3. Ghanem MA, Wolffenbuttel KP, De Vylder A and Nijman RJ: Long-term bladder dysfunction and renal function in boys with posterior urethral valves based on urodynamic findings. J Urol 2004; 171: 2409. 4. Smith GH, Canning DA, Schulman SL, Snyder HM 3rd and Duckett JW: The long-term outcome of posterior urethral valves treated with primary valve ablation and observation. J Urol 1996; 155: 1730. 5. Tejani A, Butt K, Glassberg K, Price A and Gurumurthy K: Predictors of eventual end stage renal disease in children with posterior urethral valves. J Urol 1986; 136: 857. 6. Ylinen E, Ala-Houhala M and Wikstrom S: Prognostic factors of posterior urethral valves and the role of antenatal detection. Pediatr Nephrol 2004; 19: 874. 7. Bajpai M, Dave S and Gupta DK: Factors affecting outcome in the management of posterior urethral valves. Pediatr Surg Int 2001; 17: 11.
1708
RISK FACTORS FOR END STAGE RENAL DISEASE AND POSTERIOR URETHRAL VALVES
8.
McGuire EJ, Woodside JR and Borden TA: Upper urinary tract deterioration in patients with myelodysplasia and detrusor hypertonia: a followup study. J Urol 1983; 129: 823. 9. Koff SA, Mutabagani KH and Jayanthi VR: The valve bladder syndrome: pathophysiology and treatment with nocturnal bladder emptying. J Urol 2002; 167: 291. 10. DeFoor W, Tackett L, Minevich E, McEnery P, Kitchens D, Reeves D et al: Successful renal transplantation in chil-
11.
12.
dren with posterior urethral valves. J Urol 2003; 170: 2402. Cuckow PM, Dinneen MD, Risdon RA, Ransley PG and Duffy PG: Long-term renal function in the posterior urethral valves, unilateral reflux and renal dysplasia syndrome. J Urol 1997; 158: 1004. Narasimhan KL, Mahajan JK, Kaur B, Mittal BR and Bhattacharya A: The vesicoureteral reflux dysplasia syndrome in patients with posterior urethral valves. J Urol 2005; 174: 1433.
DISCUSSION Dr. Gordon McLorie. In 1991 we published an article entitled “Nadir creatinine in posterior urethral valves.” It is either depressing or encouraging to note that the data still stand up and we have not come much further to predicting which children are going to end up with bad kidneys. Can you briefly answer what your major indications were for vesical or supravesical diversion in children with posterior urethral valves? Dr. William DeFoor. It is a difficult clinical decision. Generally, after valve ablation the serum creatinine remains high, or there is significant stasis or progressive hydronephrosis. I think in those children alone we would consider upper tract diversion but not as a primary treatment modality. Dr. Howard Snyder. I feel compelled to say one thing because a big Texan up in the sky whispered something in my ear. We fought this battle of supravesical diversion 20 years ago, and I thought it was dead. In his last paper before he died John Duckett made a plea to avoid supravesical diversion. We have had an abundance of experience with this and while vesicostomy lets a bladder continue to cycle and does not hurt it, supravesical diversion should never be done. There is no indication that a single patient has ever benefited by that over vesicostomy. I say that strongly because I think that is really the truth. There are too many young people here and I do not think they should get any mixed messages about this. Supravesical diversion prevents the bladder from cycling and if you want to cause harm to a patient with valves, do a supravesical diversion. Dr. Kenneth Glassberg. Gordon McLorie was talking about an important paper from Toronto on nadir creatinine, and other institutions have reported nadir creatinine as well. There is an important report by Kevin Burbidge who talked about continence at the age of 5. It is almost the same type of outcome because when you are continent at the age of 5 creatinine is much less. So continence and nadir creatinine seem to be important but can we affect either? I think all of these patients need urodynamics from the onset because we need to make sure that the valve is completely ablated. We should be doing videourodynamics and not VCUG. If we could affect continence by the age of 5 we will prevent renal failure in some patients or slow it down by age 10. We must be proactive. Right after valve ablation start thinking about anticholinergics because if you lower the pressures early in life, the natural tendency to develop myogenic failure does not occur. Also if you lower the pressure early, you might affect when or if end stage renal disease develops. Don’t just accept that dysplastic kidneys are going to result in end stage renal disease. We can have a positive affect. Dr. Joao Pippi Salle. I teach my fellows that occasionally there is a case for which ureterostomy or high diversion should be done based on a previous report by Jarrod Wisser published a few years ago. High diversion delayed the onset of end stage renal disease. There was no difference in bladder function and all patients had normal urodynamic parameters. Although I agree that valve ablation should be performed in most cases, high diversion should be considered in select cases. This is important for the young people to know.