Unabridged Version of this Guideline [pdf]

Index patient

The 4-year old patient with vesicoureteral reflux (VUR) and no clinical evidence of bladder/bowel dysfunction (BBD) who has presented initially with a febrile urinary tract infection (UTI) and a subsequent diagnosis of VUR by cystography.

Introduction

The association between VUR and febrile UTI was established in the 1997 AUA Guideline for Pediatric Vesicoureteral Reflux1 based upon a significantly higher incidence of febrile UTI in children with ongoing VUR than in those with resolved or surgically treated VUR. Identification and management of VUR provides the potential opportunity to prevent renal damage. Children with UTI in the setting of VUR are at an increased risk for febrile UTI and pyelonephritis. Pyelonephritic episodes can be associated with renal scarring. The odds of renal scarring in children with VUR and pyelonephritis are 2.8 times greater than the odds of scarring for children with pyelonephritis without VUR. In order to reduce the morbidity of acute pyelonephritis and the risk of permanent renal injury, treatment of VUR is recommended. Treatment options include observation, continuous antibiotic prophylaxis (CAP), and interventions of curative intent.

Methodology

Literature Search, Data Extraction, and Evidence Combination

A meta-analysis of the existing literature was performed to define the outcomes of nonoperative and surgical management of VUR among toilet-trained children without evidence of BBD. The analysis defined 19 specific questions upon which clinical decisions are based. The questions were grouped into three categories; 1) those evaluating the causal relationship between VUR, pyelonephritis and renal injury (no causal relationship can be established with the design used in the majority of studies), 2) those determining the outcomes following treatment with CAP and 3) the outcomes of therapy with curative intent (endoscopic and open surgical intervention).

In all, 205 articles were evaluable, and data from 84 studies reported from 1994 to 2008 were extracted and meta-analyzed. Some studies included children younger than one year and outcomes were not separable for younger and older children. Also, some evidence may include non-toilet trained children (>1 yr of age). Data were extracted on 11,837 children: 578 children who underwent observation in seven studies, 3,311 children who received CAP in 29 studies, 4,587 children who had open surgery in 33 studies, 626 children who received macroplastique injection therapy in 7 studies, and 2,735 children who received dextranomer/hyaluronic acid injection therapy in 17 studies.

In addition, a subanalysis was conducted to determine the possible association between acute pyelonephritis, VUR and renal scarring. A meta-analysis was performed comparing the incidence of renal scarring after the occurrence of an episode of acute pyelonephritis in children with VUR to those without VUR. The articles for the subanalysis, which were not selected in the initial literature review, included only studies in which children had acute pyelonephritis based on technetium-99m-labeled dimercaptosuccinic acid (DMSA) abnormalities and who were re-evaluated by DMSA more than 6 months after the acute event. These studies were not exclusively comprised of children with VUR. The remaining criteria used in selecting articles for this subanalyis are described in the methodology. Any study not meeting these strict criteria was excluded. Included in the analysis were eight studies of cohorts assessed between 1992 and 2006, with five (63%) studies being prospective. In all, at least 266 children with VUR and 444 without VUR were examined with a second DMSA for the development of scarring.

A summary of outcomes of the meta-analysis and relevant Guideline statements for management of VUR in the child over 1 year of age is presented below.

Initial evaluation of the Child with VUR

VUR, Acute Pyelonephritis and Renal Injury

The first set of questions assessed the association between VUR and renal injury using DMSA abnormalities, somatic growth impairment, hypertension, and renal insufficiency as indicators. Correlation of DMSA abnormality with age at diagnosis and reflux grade was evaluated; since most studies did not specify the age at diagnosis, the age at initiation of antibiotic or surgical therapy was used as a surrogate. The prevalence of renal cortical abnormalities decreased by 5.5% per yearly increase in average age at initiation of therapy (Figure 1). In studies with younger patients, the prevalence of DMSA abnormalities was higher than in studies with older children. It is unclear if this reflects a selection effect or demonstrates a greater tendency for DMSA abnormalities in younger children.

Figure 1. Relationship between renal cortical abnormalities on DMSA scanning and age at initiation of therapy

Although few studies explicitly described DMSA abnormalities by VUR severity, studies with a greater proportion of children having VUR grades I-II demonstrated a lower prevalence of renal cortical abnormalities, while a higher prevalence of renal cortical abnormalities was seen in studies of children with grades III to V VUR (Figure 2).

Figure 2. Prevalence of renal cortical abnormalities by VUR severity (by patient)

The number of studies using somatic growth, hypertension, and renal functional loss as indicators were limited. The evaluation of somatic growth impairment was assessed in one study with 94 children diagnosed with VUR following a UTI.² Children with bilateral reflux and scarring were more likely to have baseline growth impairment, starting with a height Z score of -0.45, in contrast to +0.18 in those with unilateral reflux and scarring. With either medical therapy (CAP) or surgical therapy, clear catch-up growth was demonstrated in those with bilateral reflux and scarring, with a height Z score of +0.21 after a mean of 3.1 years.³ In nine studies reporting data on hypertension, the incidence was 1 per 100 children (95% confidence interval [CI]: 0.4, 2.8). These studies were limited by a short duration of follow-up in most cases. The few studies with long-term follow-up4 suggest there is an increased incidence of hypertension over the long-term and it would seem prudent to recommend monitoring blood pressure at routine follow-up visits in children with known VUR and certainly in those with renal scarring.⁴ Similarly, the incidence of renal insufficiency in eight studies was determined to be 1.7 per 100 children followed (95% CI: 0.3, 8.2). It should be noted that definitions of functional renal loss were variable and included decreased differential function of more than 6% on radionuclide imaging, elevated serum creatinine, renal failure, and end-stage renal disease.

Association between Renal Damage and UTI

To determine whether there was an association between renal damage and UTI, three approaches were used: 1) evaluation of whether the incidence of renal damage was greater for those presenting with UTI versus those not presenting with a UTI; 2) evaluation of whether the number of prior UTIs was related to the incidence of renal injury (this could not be addressed due to a limited number of studies); and 3) evaluation of the outcomes of studies in which the incidence of new renal cortical abnormalities occurring more than 6 months after an episode of strictly defined acute pyelonephritis were correlated with the presence or absence of VUR.

Although the relationship between the initial incidence of UTI and the presence of DMSA renal cortical abnormalities in the setting of VUR may reflect the interplay of multiple factors, a higher prevalence of baseline UTI appeared to be associated with a higher prevalence of renal cortical abnormalities (Figure 3).

Figure 3. Relationship between baseline UTI and DMSA renal cortical abnormalities

To determine whether children with VUR are at greater risk of having permanent renal injury following acute pyelonephritis compared to those without VUR, the incidence of both acute and permanent DMSA cortical abnormalities in children with and without VUR were analyzed. A 2.8 fold increase in the risk of permanent DMSA cortical abnormalities in the children with VUR and a 3.7 fold higher risk for individual renal units were found (Figure 4), showing that VUR is a significant risk factor in the development of acute pyelonephritis and subsequent renal damage in children.

Figure 4. Forest plots of odds ratios on a log scale of scarring after acute pyelonephritis among children with VUR compared to those without VUR. A) by patient; B) by renal unit.

Standard: VUR and urinary tract infections may detrimentally affect the overall health and renal function in affected children. Therefore, on initial presentation the child with VUR should undergo a careful general medical evaluation including measurement of height, weight, and blood pressure, and serum creatinine if bilateral renal abnormalities are found.

[Based on Panel consensus]

Recommendation: Urinalysis for proteinuria and bacteriuria. If the urinalysis indicates infection, a urine culture and sensitivity is recommended.

[Based on Panel consensus]

Recommendation: Because VUR and urinary tract infection may affect renal structure and function, performing renal ultrasound to assess the upper urinary tract is recommended.

[Based on Panel consensus]

Option: A baseline serum creatinine may be obtained to establish an estimate of glomerular filtration rate (GFR) for future reference.

[Based on Panel consensus]

Option: DMSA (technetium-99m-labeled dimercaptosuccinic acid) renal imaging can be obtained to assess the status of the kidneys for scarring and function.

[Based on review of the data and Panel Consensus]

Continuous Antibiotic Prophylaxis in the Treatment of VUR

Current standard therapy for VUR includes the use of CAP to prevent acute infection with the anticipation that spontaneous resolution of VUR will occur in a significant proportion of children.¹ Outcomes of treatment with CAP were evaluated in terms of incidence and character (cystitis, febrile, nonspecified) of UTI, incidence of new renal cortical abnormalities and the resolution of VUR.

UTI Incidence

Results of analysis of the incidence of UTI in children receiving or not receiving CAP in 18 studies are presented in Table 1.

Table 1. UTI incidence in patients with VUR receiving or not receiving CAP.

Note: N= number of study groups

Although the incidence of cystitis and nonspecified UTIs was similar between those receiving or not receiving CAP, the incidence of febrile UTIs in children receiving CAP was greater than those not receiving CAP. The overall incidence of cystitis and febrile UTI in those receiving or not receiving CAP, as well in the incidences from individual reports, is shown in Figure 5.

Figure 5. Incidence of A) cystitis and B) febrile UTI with CAP (filled) or without CAP (open)

Incidence of New Renal Cortical Abnormalities

The incidence of new renal cortical abnormalities identified by scintigraphy after initiation of antibiotic therapy (including those receiving CAP) was 17.1 per 100 children (95% CI: 8.6, 31.1) with VUR (Figure 6). Data regarding new renal scarring in children with VUR treated without CAP were limited.

Figure 6. Incidence of new renal cortical abnormalities during follow-up for VUR

Resolution of VUR

The overall resolution rate with CAP was 53.2 per 100 children and 67.8 per 100 children (95% CI: 56.8, 77.1) in studies with 25% or fewer males, with the highest rates occurring between 24 and 36 months (range of 12–71 months). At this time, the published resolution curves from the 1997 AUA Guidelines¹ represent the most accurate estimates of resolution; however since few studies have assessed the impact of BBD on resolution, it is not possible to determine whether VUR resolves more slowly in children with BBD.

Option: Continuous antibiotic prophylaxis may be considered for the child with a history of urinary tract infection and VUR in the absence of bladder/bowel dysfunction.

[Based on review of the data and Panel consensus]

Option: Observational management without continuous antibiotic prophylaxis, with prompt initiation of antibiotic therapy for urinary tract infection, may be considered for the child with VUR in the absence of bladder/bowel dysfunction, recurrent febrile urinary tract infectoins, or renal cortical abnormalities.

[Based on Panel consensus]

Follow-up Management of the Child with VUR

Ongoing monitoring of a child's overall health is necessary. Specific testing related to VUR will depend on the clinical situation and any factors described below that might indicate the potential for ongoing or progressive renal injury.

General

Recommendation: General evaluation, including monitoring of blood pressure, height, and weight is recommended annually.

[Based on Panel consensus]

Recommendation: Urinalysis for proteinuria and bacteriuria is indicated annually, including a urine culture and sensitivity if the urinalysis is suggestive of infection.

[Based on Panel consensus]

Option: The follow-up interval is determined by the likelihood of resolution; for higher grades of VUR, resolution is less likely¹ and therefore a longer interval of follow-up is appropriate.

[Based on Panel consensus]

Imaging

Recommendation: Imaging, including ultrasonography and voiding cystography (radionuclide cystogram or low-dose fluoroscopy, when available) is recommended every 12 to 24 months, with longer intervals between follow-up studies in patients in whom evidence supports lower rates of spontaneous resolution to limit the number of imaging studies (i.e. those with higher grades of VUR (grades III-V; see clinical chapter 2 and 1997 Guidelines), bladder/bowel dysfunction (see clinical chapter 3), and older age). This is to limit the over-all number of imaging studies performed. If an observational approach is being used, follow-up cystography becomes an option.

[Based on review of the data and Panel consensus]

Recommendation: DMSA imaging is recommended when a renal ultrasound is abnormal, when there is a greater concern for scarring (i.e. breakthrough urinary tract infectoin [BT-UTI; see Glossary for description], higher grades III-V of VUR), or if there is an elevated serum creatinine.

[Based on review of the data and Panel consensus]

Option: Follow-up cystography may be done after one year of age in patients with VUR grades I–II; these patients tend to have a high rate of spontaneous resolution and boys have a low risk of recurrent urinary tract infection.

[Based on review of the data and Panel consensus]

Option: A single normal voiding cystogram (i.e. no evidence of VUR) may serve to establish resolution. The clinical significance of grade I VUR, and the need for ongoing evaluation is undefined.

[Based on review of the data and Panel consensus]

Option: Periodic upper tract imaging with renal ultrasound may be done to assess renal growth and the presence of gross renal scarring.

[Based on Panel consensus]

Option: DMSA may be considered for follow-up of children with VUR, to detect new renal scarring, especially after a febrile urinary tract infection.

[Based on review of the data and Panel consensus]

Interventions for the child with a febrile breakthrough UTI

When a febrile breakthrough UTI (BT-UTI) occurs in a child with VUR receiving CAP, consideration of alternative interventions is recommended since there is the potential for renal injury. The clinical manifestations of BT-UTI may not be classic, particularly in the younger child in whom systemic symptoms may predominate.

Recommendation: If symptomatic breakthrough urinary tract infection occurs (manifested by fever, dysuria, frequency, failure to thrive, or poor feeding), a change in therapy is recommended. If symptomatic breakthrough urinary tract infection occurs, the clinical scenario will guide the choice of treatment alternatives; this includes VUR grade, degree of renal scarring, if any, and evidence of abnormal voiding patterns (bladder/bowel dysfunction) that might contribute to urinary tract infection, and parental preferences.

[Based on Panel consensus]

Recommendation: It is recommended that in patients receiving continuous antibiotic prophylaxis with a febrile breakthrough urinary tract infection be considered for open surgical ureteral reimplantation or endoscopic injection of bulking agents for intervention with curative intent.

[Based on Panel consensus]

Recommendation: In patients not receiving continuous antibiotic prophylaxis who develop a febrile urinary tract infection, initiation of continuous antibiotic prophylaxis is recommended.

[Based on Panel consensus]

Option: In patients receiving continuous antibiotic prophylaxis with a single febrile breakthrough urinary tract infection and no evidence of pre-existing or new renal cortical abnormalities, changing to another antibiotic agent is an option prior to intervention with curative intent.

[Based on Panel consensus]

Option: In patients not receiving continuous antibiotic prophylaxis who develop a non-febrile urinary tract infection, initiation of continuous antibiotic prophylaxis is an option in recognition of the fact that not all cases of pyelonephritis are associated with fever.

[Based on Panel consensus]

Surgical treatment of VUR

Outcomes of open and endoscopic surgical approaches including cure rate, incidence of UTI, renal cortical abnormalities, and adverse events were assessed. Twenty-nine studies evaluated the management of VUR with open surgery, 18 of which reported resolution rates. The resolution rate per 100 children was 98.1 for open surgery (95% CI: 95.1, 99.1) and 83.0 for endoscopic therapy (95% CI: 69.1, 91.4) after a single injection of bulking agent. These studies typically reported reflux status in the first year after intervention. Data and clinical experience demonstrating the durability of endoscopic therapy for VUR are limited. The novelty of the treatment has not permitted adequate assessment of its long-term efficacy in a broad range of patients. While few studies have directly assessed the long-term durability of open surgery for VUR, clinical experience indicates that recurrence is rare.⁵

UTI after surgical therapy

Twenty-four studies assessed the incidence of UTI after either endoscopic or open correction of VUR. The incidence of postoperative febrile UTI was 4.9 cases per 100 children (95% CI: 2.2, 10.4). The rate of postoperative cystitis was higher in the operative group (15%) versus those receiving CAP (7.2%) and not receiving CAP (7.9%) groups. Higher preoperative UTI rates were associated with higher postoperative UTI rates. In studies in which fewer than 60% of children had preoperative UTIs, the postoperative UTI incidence was 4.6 per 100 children (95% CI: 2.2, 9.6). In studies in which more than 60% of children had preoperative UTIs, the postoperative UTI incidence was 10.2 per 100 children (95% CI: 4.0, 26.2) (Figure 7). It is important to recognize that in the absence of randomized trials controlled for preoperative UTI incidence comparisons of outcomes in patients with preoperative and postoperative UTIs may not be reliable.

Figure 7. Incidence of postoperative UTI in relation to the incidence of baseline UTI prevalence in the study population (ecological association)

New renal cortical abnormalities after surgical therapy

The 1997 AUA Guideline1 found the incidence of new renal cortical abnormalities occurring after surgical versus CAP therapy to be similar; limited useful new data have been reported. In this analysis, the incidence of renal cortical abnormalities after surgery was assessed in 450 patients in four studies with a broad range of 3.0%–38.3%. Due to this extreme variability, which may reflect differences in selection criteria for surgical intervention, it is not possible to determine an overall assessment of the incidence of new renal cortical abnormalities after surgical therapy.

The number of adverse events following endoscopic or open surgery for VUR was low. The overall postoperative obstruction rate calculated from 28 articles was 0.4 (95% CI: 0.2, 1.2) per 100 children. The incidence of postoperative voiding disturbances was 4.2 per 100 children (95% CI: 1.8, 9.2) while the incidence of postoperative contralateral VUR after unilateral treatment was 9.6 per 100 children (95% CI: 7.5, 12.2).

Option: Surgical intervention for VUR, including both open and endoscopic methods, may be used. Prospective randomized controlled trials (RCTs) have shown a reduction in the occurrence of febrile urinary tract infections in patients who have undergone open surgical correction of VUR as compared to those receiving continuous antibiotic prophylaxis.^{5, 6}

[Based on review of the data and Panel consensus]

Management following resolution of VUR

Although there are no data with which to assess a specific follow-up program, it is recognized that the presence renal injury is associated with a higher risk of later health effects. While these risks may be low, they are known to increase with the duration of follow-up.

Option: Following the resolution of VUR, either spontaneously or by surgical intervention and if both kidneys are normal by ultrasound or DMSA scanning, general evaluation, including monitoring of blood pressure, height, and weight, and urinalysis for protein and urinary tract infection, annually through adolescence is an option.

[Based on Panel consensus]

Recommendation: Following the resolution of VUR, either spontaneously or by surgical intervention, general evaluation, including monitoring of blood pressure, height, and weight, and urinalysis for protein and urinary tract infection, is recommended annually through adolescence if either kidney is abnormal by ultrasound or DMSA scanning.

[Based on Panel consensus]

Recommendation: With the occurrence of a febrile urinary tract infection following resolution or surgical treatment of VUR, evaluation for bladder/bowel dysfunction or recurrent VUR is recommended.

[Based on Panel consensus]

Recommendation: It is recommended that the long-term concerns of hypertension (particularly during pregnancy), renal functional loss, recurrent urinary tract infection, and familial VUR in the child's siblings and offspring be discussed with the family and communicated to the child at an appropriate age.

[Based on Panel consensus]

Summary

An association between VUR and renal injury was demonstrated in this meta-analysis after reviewing studies evaluating renal cortical abnormalities on DMSA scanning at diagnosis and examination of specific endpoints secondarily related to renal damage such as hypertension, renal insufficiency, and somatic growth impairment. These observations support therapeutic management of VUR in children in order to reduce the risk of pyelonephritis and potential permanent renal injury.

Spontaneous resolution of VUR does occur and remains the basis for expectant management where CAP is used to prevent potentially injurious UTI. The efficacy of CAP in preventing febrile UTI or renal injury remains uncertain. There have been few studies comparing the outcomes of CAP in children with VUR until recently and the generalizability of these studies to the broader patient population has not been established. One retrospective study showed no difference in the occurrence of UTI or scarring after discontinuing CAP.⁷ Three recently published studies have addressed the question of the clinical benefit of CAP in children with VUR.^8-10 All were prospective, multicenter, randomized studies in children with their first UTI. In the study by Garin et al.,⁸ the wide range of ages (from 1 month to 18 years) and the inability to define specific outcomes for the subgroups with the criteria used to assess other reports limited its utility; it was therefore not included in this meta-analysis. For studies included in this meta- analysis, Roussey-Kessler et al.¹⁰ showed a benefit for CAP in boys with grade III VUR but did not address higher grades while Pennesi et al.⁹ included children with grade IV VUR, but in limited numbers. In these studies there was no assessment of voiding patterns and medication compliance, there was significant heterogeneity in the patient population raising questions about the variation in selection criteria, and in one study UTIs were documented by bag urines. At this point, it is uncertain whether there is a benefit to the use of CAP in children diagnosed with VUR following a first febrile UTI or for patients with VUR grades III, IV and V.

The Cochrane analyses of the use of CAP in treatment of VUR have also been cited as demonstrating the lack of benefit with the use of CAP.^{11, 12} These systematic reviews have served as the basis for the recent publication of the National Institute for Health and Clinical Excellence Guidelines related to evaluation and management of UTI in children for the National Health Service in Great Britain.¹³ However, the Cochrane systematic reviews include only RCTs, which is a major limitation since very few studies are included.

From a practical standpoint, a cautious approach should be taken in the management of children with low-grade VUR pending the availability of data from recently reported RCTs¹⁴ and including the Swedish Reflux Study (to be published in the Journal of Urology in 2010; reported at the International Conference on Vesicoureteral Reflux in Children, Goteborg, 2009), and the Randomized Intervention for Children with Vesicoureteral Reflux [RIVUR] study (www.rivur.net)). Thus our Guideline includes the option of managing children without the use of CAP while maintaining careful observation. Continuing the use of CAP in children with grades II-V VUR seems prudent until more data are available.

If there is indeed a lack of preventative benefit with CAP for acute pyelonephritis and renal injury in children with VUR, one may question the value of treating, or even diagnosing, VUR. Alternately, the implication may be that every child with VUR should undergo surgical intervention using open or endoscopic methods. Neither can be justified as yet in routine clinical practice for all patients. Selection criteria have not been defined nor validated for those patients in whom a non-interventional approach is appropriate. It would seem imprudent to discard an approach that evolved from careful clinical observation and shows no evidence of harm, until more robust data are available in terms of the certainty of the safety of non-interventional management of VUR, and in whom this approach is appropriate.

Observational management without the administration of CAP (with prompt initiation of antibiotic therapy for acute UTI), is currently under investigation, so no recommendations regarding this form of therapy can be made at this time. Decisions regarding choice of treatment depend on a number of factors such as patient age, VUR grade, the presence of scarring at diagnosis, and parental preferences. The likelihood of resolution of VUR by observational, endoscopic, and open surgical therapy should be factored into the decision-making process. Higher VUR grades and the presence of scarring would more strongly favor a curative intervention to limit risk of further damage in a child with reduced renal reserve.

There are no data assessing the value of any particular follow-up regimen in children with ongoing or resolved VUR. Any Guideline must be based on a clinical judgment as to the potential for late effects of renal injury and UTI, and the ability to intervene if these effects occur. The principal concerns include the development of hypertension and renal insufficiency, as well as acute pyelonephritis from cystitis or bacteriuria. The few long-term studies on the impact of scarring, albeit in a selected population, include that of Smellie et al.,⁴ in which the incidence of late hypertension, renal insufficiency and complications of pregnancy were associated with the severity of scarring and the duration of follow-up, and of Martinell et al.,¹⁵ in which pyelonephritis during pregnancy was related to renal scarring and the prior incidence of UTI. It should be noted that the National Heart, Lung and Blood Institute and the American Academy of Pediatrics recommend that a blood pressure determination should be made in all children over the age of 3 years.¹⁶ This recommendation is for all children, not just those with demonstrable renal injury related to VUR, and is applicable to any child with major illness, including pyelonephritis.

References

1. Elder, J. S., Peters, C. A., Arant, B. S., Jr. et al.: Pediatric Vesicoureteral Reflux Guidelines Panel summary report on the management of primary vesicoureteral reflux in children. J Urol 1997; 157: 1846.

2. Polito, C., La Manna, A., Mansi, L. et al.: Body growth in early diagnosed vesicoureteric reflux. Pediatr Nephrol 1999; 13: 876.

3. Polito, C., Marte, A., Zamparelli, M. et al.: Catch-up growth in children with vesico-ureteric reflux. Pediatr Nephrol 1997; 11: 164.

4. Smellie, J. M., Prescod, N. P., Shaw, P. J. et al.: Childhood reflux and urinary infection: a follow-up of 10-41 years in 226 adults. Pediatr Nephrol 1998; 12: 727.

5. Jodal, U., Smellie, J. M., Lax, H. et al.: Ten-year results of randomized treatment of children with severe vesicoureteral reflux. Final report of the International Reflux Study in Children. Pediatr Nephrol 2006, 21: 785.

6. Jodal, U., Koskimies, O., Hanson, E. et al.: Infection pattern in children with vesicoureteral reflux randomly allocated to operation or long-term antibacterial prophylaxis. The International Reflux Study in Children. J Urol 1992, 148: 1650.

7. Thompson, R. H., Chen, J. J., Pugach, J. et al.: Cessation of prophylactic antibiotics for managing persistent vesicoureteral reflux. J Urol 2001, 166: 1465.

8. Garin, E. H., Olavarria, F., Garcia Nieto, V. et al.: Clinical significance of primary vesicoureteral reflux and urinary antibiotic prophylaxis after acute pyelonephritis: a multicenter, randomized, controlled study. Pediatrics 2006, 117: 626.

9. Pennesi, M., Travan, L., Peratoner, L. et al.: Is antibiotic prophylaxis in children with vesicoureteral reflux effective in preventing pyelonephritis and renal scars? A randomized, controlled trial. Pediatrics 2008, 121: e1489.

10. Roussey-Kesler, G., Gadjos, V., Idres, N. et al.: Antibiotic prophylaxis for the prevention of recurrent urinary tract infection in children with low grade vesicoureteral reflux: results from a prospective randomized study. J Urol 2008, 179: 674.

11. Hodson, E. M., Wheeler, D. M., Vimalchandra, D. et al.: Interventions for primary vesicoureteric reflux. Cochrane Database Syst Rev 2007: CD001532.

12. Wheeler, D. M., Vimalachandra, D., Hodson, E. M. et al.: Interventions for primary vesicoureteric reflux. Cochrane Database Syst Rev 2004: CD001532.

13. NCCWCH: Urinary Tract Infection in Children - diagnosis, treatment and long- term management. London: RCOG Press, p. 148, 2007

14. Craig, J. C., Simpson, J. M., Williams, G. J. et al.: Antibiotic prophylaxis and recurrent urinary tract infection in children. N Engl J Med 2009, 361: 1748.

15. Martinell, J., Claesson, I., Lidin-Janson, G. et al.: Urinary infection, reflux and renal scarring in females continuously followed for 13-38 years. Pediatr Nephrol 1995, 9: 131.

16. National High Blood Pressure Education Program Working Group on High Blood Pressure in Children and, A.: The Fourth Report on the Diagnosis, Evaluation, and Treatment of High Blood Pressure in Children and Adolescents. Pediatrics 2004, 114: 555.

Unabridged Version of this Guideline [pdf]

Index patient

The infant less than one year of age who is diagnosed with primary vesicoureteral reflux (VUR) during the early postnatal period based on a diagnosis of prenatal hydronephrosis (PNH) or following the occurrence of a urinary tract infection (UTI).

Introduction

Vesicoureteral reflux in infants is usually diagnosed after a febrile UTI or during the postnatal work-up of a child with PNH. To prevent any detrimental impact on long-term renal function, early detection of a febrile UTI is critical in infants, who are unable to verbally communicate lower urinary tract symptoms. The management of infants with VUR has become increasingly controversial due to systematic reviews of the literature which support the protective role of circumcision and studies that question the time-honored value of prescribing continuous antibiotic prophylaxis (CAP). Moreover, there has been a relatively recent paradigm shift following the introduction of early treatment by endoscopic injection therapy. Management of infants with VUR should take into consideration the likelihood of spontaneous resolution, the likelihood of recurrence of UTI and the risk of developing renal parenchymal abnormalities.

Methodology

Literature Search, Data Extraction, and Evidence Combination

A meta-analysis of the existing literature was performed to determine the effects of nonoperative management (i.e., CAP) in children with VUR who are less than one year of age. Outcomes included the rates of VUR resolution, the incidence of UTI, and the incidence of renal cortical abnormalities. For inclusion, infants must have been diagnosed with VUR by cystography at or before one year of age and VUR resolution must have been assessed by at least one follow-up cystogram. Assessment by renal scintigraphy (technetium-99m-labeled dimercaptosuccinic acid, diethylenetriamine pentaacetate, or mercaptoacetyltriglycine) was required for studies reporting data on renal cortical abnormalities.

Twenty-one studies met the inclusion criteria (six were prospective), data were extracted and a meta-analysis was performed. Of these, four provided data on VUR resolution by sex, nine provided data on resolution rates by VUR grade (based on renal units), seven provided data on renal cortical abnormalities, and 15 provided data on occurrence of UTI (with varying criteria for UTI diagnosis) in patients receiving CAP. These reports included data on 1,323 infants managed with CAP; 81.0% had a diagnosis of PNH and 67.8% were male.

Outcomes Analysis

Vesicoureteral Reflux Resolution

The resolution rate for infants with PNH or UTI at presentation was 49.9 per 100 patients and 52.0 per 100 renal units (Figure 1), with the range of follow-up times between 12 and 48 months. The resolution rate for studies with infants diagnosed prenatally (no prior UTI) was 59.2 per 100 infants, representing a 20% greater VUR resolution compared to the rate for all infants.

Figure 1. Forest plot of VUR resolution rates among infants (A) per patient and (B) per renal unit.

When stratified by sex, the resolution rates were 46.8 for males and 51.6 for females per 100 renal units (a nonsignificant difference). Renal units with severe grade (IV–V) VUR were 85% less likely to resolve than those with mild-moderate grade VUR (71/100 units with grades I–III VUR vs. 28/100 units with grades IV–V VUR; p-value<0.0001).

Incidence of Urinary Tract Infection

The incidence of UTI in infants diagnosed with PNH or UTI who were managed with CAP was 19.5 cases per 100 infants, with a mean follow-up of 24 months (Figure 2); the incidence of UTI for infants with PNH and no prior UTI was 17.5 per 100. The effect of gender, VUR grade or circumcision status could not be analyzed due to the insufficient data. Similarly, there was a lack of data clearly separating patients with proven pyelonephritis.

Figure 2. Forest plot of UTI incidence in infants with VUR managed with continuous antibiotic prophylaxis.

Among infants initially receiving CAP, 17.7% had subsequent surgical treatment, usually with an open procedure. Criteria used to determine the need for surgery included persistent grade III-V VUR, development of a breakthrough UTI (BT-UTI), treatment noncompliance and deterioration of renal function. The distribution of VUR severity across children in the individual samples did not explain the rate of surgery, indicating that other factors were involved in the choice to proceed to surgical repair.

Renal Cortical Abnormalities

The overall incidence of renal cortical abnormalities (scarring or decreased uptake on scintigraphy) was 8.2 per 100 infants with VUR. It is not clear how many of these defects were present since birth (or without prior pyelonephritis), thereby representing congenital changes rather than scarring secondary to UTI. Due to the relatively small sample size, the potential effect of persistent VUR, high-grade VUR or UTI on renal outcomes could not be analyzed. Paucity of data on renal cortical abnormalities, as detected by scintigraphy, limits the analysis of heterogeneity and the conclusions that can be reached on this important outcome.

Recommendation: Continuous antibiotic prophylaxis is recommended for the child less than one year of age with VUR with a history of a febrile urinary tract infection. This approach is based on the greater morbidity from recurrent urinary tract infection found in this population.

[Based on review of the data and Panel consensus]

Recommendation: In the absence of a history of febrile urinary tract infection, continuous antibiotic prophylaxis is recommended for the child less than one year of age with VUR grades III–V who is identified through screening.

[Based on review of the data and Panel consensus]

Option: In the absence of a history of febrile urinary tract infection, the child less than one year of age with VUR grades I–II who is identified through screening may be offered continuous antibiotic prophylaxis.

[Based on review of the data and Panel consensus]

Option: Circumcision of the male child with VUR may be considered based on an increased risk of urinary tract infection in boys who are not circumcised compared to those who are circumcised. Although there are insufficient data to evaluate the degree of this increased risk and its duration, parents need to be made aware of this association to permit informed decision-making.

[Based on review of the data and Panel consensus]

Summary

Primary VUR that is diagnosed during infancy and managed nonoperatively with CAP is likely to resolve within 12–48 months in approximately half of all children. This is an aggregate percentage that includes patients with a diagnosis of PNH as well as those diagnosed with a UTI. The probability of resolution was influenced by the initial VUR grade (a lower VUR grade had a higher resolution rate), a finding that supports the value of considering VUR grade when counseling parents. The reported UTI incidence data appeared to reflect the era in which they were published, with earlier studies (before the 1990s) tending to report lower incidences than those more recently published. These differences may reflect, among many things, better reporting, a heightened level of suspicion (and subsequent evaluation), and/or changes in the pattern of drug resistance. Not surprisingly, many outliers with a low reported incidence of UTI had an average follow-up of less than 18 months. This highlights the need for longer follow-up of events for which increased occurrence is likely to be detected with longer observation.

The infant with VUR may have presented with prenatally detected hydronephrosis that prompted performance of the voiding cystourethrogram, or they may have presented with a UTI. It is widely presumed, without evidence, that these different patients may be distinct in terms of natural history and further risk of UTI and renal injury. The Panel attempted to assess this using the available data, but could not identify sufficient numbers of studies in which patient outcomes were stratified by presentation. The incidence of UTI is slightly different between those presenting after prenatal detection and after UTI (Fig. 2). However, we could only separate out a subset of children with just PNH from the group, including both UTI and PNH presentation. We were not able to compare those presenting as UTI with those presenting with PNH. Therefore the data are presented as an aggregate group. No evidence is available that either supports or refutes the idea that infants with VUR presenting after UTI are distinct from those presenting with PNH in terms of risk of renal injury. A cautious approach would be to consider both groups as being at risk for further UTI until proven otherwise.

There are limited data to assess the efficacy of CAP in this subgroup of children. In the prospective trials, Roussey-Kesler, et al.¹ reported a benefit for boys with grade III VUR, while Pennesi, et al.² found no benefit for infants; however, both of these trials had very limited numbers of patients in these groups, suggesting that generalizing these findings to infants is risky. Most of the reviewed studies were conducted in an era when CAP was the standard of care; in the absence of a control group not receiving CAP, the role of CAP in VUR management cannot be accurately assessed. Although these recent studies of varying methodological quality raise questions as to its benefit, until its efficacy is specifically assessed in this population, CAP is recommended to protect infants while awaiting spontaneous resolution.

The role of circumcision in male patients, a potentially important variable in terms of UTI risk, could not be determined from the reviewed literature. While circumcision may have an important impact on the risk of UTI in boys³, its specific effect in the context of VUR could not be adequately assessed due to the lack of available data regarding circumcision status in the evaluated populations. The established reduction in UTI, however, coupled with the risk of UTI in infants with VUR, prompted the panel to consider circumcision an option in the management of the infant boy with VUR.

In this meta-analysis, upper urinary tract evaluation showed renal cortical abnormalities in approximately 14% of patients, including those without prior UTIs; it was not possible to distinguish new abnormalities (renal scarring) from defects present since birth (congenital renal abnormality). With longer term follow-up, patients at risk for further damage may be detected.

References

1. Roussey-Kesler, G., Gadjos, V., Idres, N. et al.: Antibiotic prophylaxis for the prevention of recurrent urinary tract infection in children with low grade vesicoureteral reflux: results from a prospective randomized study. J Urol 2008; 179: 674.

2. Pennesi, M., Travan, L., Peratoner, L. et al.: Is antibiotic prophylaxis in children with vesicoureteral reflux effective in preventing pyelonephritis and renal scars? A randomized, controlled trial. Pediatrics 2008; 121: e1489.

3. Singh-Grewal, D., Macdessi, J., Craig, J.: Circumcision for the prevention of urinary tract infection in boys: a systematic review of randomised trials and observational studies. Arch Dis Child 2005; 90: 853.

Unabridged Version of this Guideline [pdf]

The 4-year-old child with vesicoureteral reflux (VUR) and evidence of clinical bladder/bowel dysfunction (BBD) without evidence of an overt neurological cause.

Introduction

Bladder/bowel dysfunction, dysfunctional voiding, dysfunctional elimination syndrome and dysfunctional lower urinary tract symptoms refer to a common but poorly characterized complex of symptoms typically including urinary incontinence, dysuria, urinary tract infections (UTI), urinary frequency, infrequent voiding and constipation. In this Guideline, BBD is used to describe children with abnormal lower urinary tract symptoms of storage and/or emptying which include lower urinary tract conditions such as overactive bladder and urge incontinence, voiding postponement, underactive bladder, and voiding dysfunction, and may also include abnormal bowel patterns including constipation and encopresis. The age of onset varies; while BBD occurs most often in the immediate post-toilet training years, it may be seen prior to and well after toilet training. While the causes are variable and not well defined, the most common pattern is felt to be due to failure of relaxation of the external sphincter and/or pelvic floor muscles leading to high voiding pressures and incomplete bladder emptying. A similar pattern may be the basis for constipation as well. The result of incomplete emptying of the bladder may predispose to the child to UTI. This, in association with high pressure voiding, may be a significant contributory factor to the natural history and health impact of VUR. The association between these entities has not been clearly defined, but several reports have shown a possible link, generally demonstrating that children with BBD have an increased risk of UTI and decreased rate of spontaneous resolution of VUR.

The appropriate approach to the management of the child with VUR and BBD has not been defined, yet the child with this combination of conditions may be at greater risk of renal injury due to infection. The questions to be asked in this context are whether BBD changes the natural history of VUR relative to that of children without BBD and whether the child with BBD is at a higher risk of renal injury or reduced success with medical or surgical management. It is important to recognize that any evidence-based review is inherently limited by the absence of any standardized description or grading system of BBD as well as the lack of uniform validated therapeutic interventions.

Methodology

Literature Search, Data Extraction, and Evidence Combination

A meta-analysis of the existing literature was performed to determine the impact of BBD on VUR in children undergoing nonoperative or surgical management in conjunction with management for elimination symptoms. Outcomes included the resolution of VUR, the incidence of UTI recurrence in children receiving medical management, and open surgical intervention subsequent to medical or endoscopic management, and the incidence of renal cortical scarring. Twelve articles described comparative studies (five were prospective) of VUR in children with or without BBD. Most children diagnosed with VUR by cystography had a previous UTI.

In all, data from 15 articles with 16 arms published between 1997 and 2006 were extracted and meta-analyzed. These reports included 2,039 children with VUR, 75% of whom were female and 640 who had a concurrent diagnosis of BBD. In children with VUR and BBD, 387 received medical management, 114 had endoscopic injection therapy and 139 had open surgery. Of those without BBD, 398 received medical management, 501 had endoscopic injection therapy, and 500 had open surgery. The mean duration of follow-up was 2.5 years for patients receiving continuous antibiotic prophylaxis (CAP) and 1 year for patients treated with endoscopic injection therapy or open surgery.

The treatments for BBD identified in the literature included bladder retraining (timed voiding, relaxed voiding, biofeedback) with or without pharmacologic (anticholinergic) intervention directed at decreasing bladder overactivity, and/or management of constipation (e.g. stool softeners). A wide range of interventions was described and there was no reliable means to compare outcomes by treatment; consequently, the results are considered in the aggregate.

Outcomes Analysis

Vesicoureteral Reflux Resolution

In children receiving CAP, resolution rates were 31% for those with BBD and 61% for those without BBD (Figure 1).

Figure 1. Forest plots of reflux resolution among children receiving continuous antibiotic prophylaxis

(Concurrent/prior use of: *bladder training, †anticholinergics, ‡stool softerners)

In children treated with endoscopic surgery resolution rates at initial follow-up were 50% for those with BBD and 89% for those without BBD (Figure 2) For children treated with open surgery, the presence of BBD did not appear to alter surgical resolution rates, which were 97% in both groups (Figure 2). In each of these groups the specific means of treating BBD and its efficacy cannot be assessed.

Figure 2. Forest plots of reflux resolution in children undergoing intervention with curative intent (open surgery or endoscopic)

Four main criteria were generally used to recommend surgery following CAP or endoscopic surgery: persistent high-grade reflux, breakthrough UTI, treatment noncompliance and/or deterioration of renal function. The time to resolution in those receiving medical management ranged from 3 to 71 months in the BBD group. No data were available regarding the specific treatments for BBD and the time to resolution of VUR. The relationship between VUR grade and time to resolution could not be assessed in either group.

The rate of open surgical correction in children who had previously been receiving CAP was 16.0 (95% confidence interval [CI]: 5.1, 40.3) per 100 cases. In children receiving CAP or undergoing endoscopic therapy, the open surgery rate was 8.3 (95% CI 2.1, 27.6) per 100 cases. Of those children initially receiving CAP, open surgical correction was performed at a mean of 30 months after initiation of CAP. A high risk of recurrent UTI, perhaps based on a history of more frequent UTI's, was the most frequent reason for surgical intervention, followed by lack of resolution with observation.

Incidence of Urinary Tract Infection

Eight studies provided data on the incidence of UTI during follow-up for 918 children with VUR, of whom 374 had concurrent BBD. The incidence of UTI managed with CAP was 44.0% (95% CI: 17.1, 74.9) in children with BBD and 12.9% (95% CI: 2.9, 42.0) in those without BBD. A large variation existed in the overall estimate of UTI in patients with BBD managed medically compared to those without BBD (Figure 3).

Figure 3. Forest plots of UTI incidence in children receiving medical management (BBD vs. non-BBD)

*bladder training; †anticholinergics; ‡stool softeners

Following open or endoscopic surgery, the incidence of UTI during follow-up was 22.6% for children with BBD and 4.8% for children without BBD (a nonsignificant difference) (Figure 4). BBD outcomes were not stratified by sex, grade or laterality for the BBD and non-BBD groups. Renal function was not specifically assessed, representing the main limitation of this analysis.

Figure 4. Forest plots of UTI incidence in children following open or endoscopic surgery (BBD vs non-BBD)

Renal Cortical Abnormalities

Five studies reported on renal outcomes in children, with two of these studies specific for children with BBD and VUR. Renal cortical abnormalities at baseline ranged from 25% to 30% in those children screened with DMSA (technetium-99m-labeled dimercaptosuccinic acid) scintigraphy. These rates are about twice those found for neonates/infants managed with antibiotics (see Chapter 2) and those in the asymptomatic siblings (see Chapter 4). There were insufficient data to determine the impact of BBD management on the rate of progression of renal scarring or the risk of new renal cortical abnormalities.

Standard: Symptoms indicative of bladder/bowel dysfunction should be sought in the initial evaluation, including urinary frequency and urgency, prolonged voiding intervals, daytime wetting, perineal/penile pain, holding maneuvers (posturing to prevent wetting), and constipation/encopresis.

[Based on Panel consensus]

Recommendation: If clinical evidence of bladder/bowel dysfunction is present, treatment of bladder/bowel dysfunction is indicated, preferably before any surgical intervention for VUR is undertaken. There are insufficient data to recommend a specific treatment regimen for bladder/bowel dysfunction, but possible treatment options include behavioral therapy (see Glossary for description), biofeedback (appropriate for children more than age five), anticholinergic medications, alpha blockers, and treatment of constipation. Monitoring the response to bladder/bowel dysfunction treatment is recommended to determine whether treatment should be maintained or modified.

[Based on Panel consensus]

Recommendation: Continuous antibiotic prophylaxis is recommended for the child with bladder/bowel dysfunction and VUR due to the increased risk of urinary tract infection while bladder/bowel dysfunction is present and being treated.

[Based on review of the data and Panel consensus]

Summary

Based on the results of this meta-analysis, children with BBD have a lower spontaneous VUR resolution rate, an increased risk of UTI (both before and after surgical correction), and a lower rate of correction after endoscopic surgery. The success rates of VUR correction with an open approach were equivalent regardless of BBD status. The incidence of baseline renal cortical abnormalities in children with BBD was higher than in infants and young children with VUR but the risk of new renal cortical abnormalites based on BBD status and/or treatment modality of VUR could not be reliably assessed.

Due to the increased rate of UTI associated with BBD, children with VUR should undergo an assessment of voiding and bowel habits. Abnormal voiding patterns should be treated in addition to preventative treatment of UTI. Due to the higher rate of UTI in children with BBD in association with VUR, CAP is recommended at least until the voiding patterns have improved. Correction of BBD should also be undertaken prior to surgical therapy unless clinical conditions are such that intervention is considered imperative, particularly in those with uncontrolled breakthrough UTIs or progressive renal damage. In that setting, open surgical correction appears to have the greatest likelihood of a cure.

Unabridged Version of this Guideline [pdf]

Index Patient

The sibling of a child with vesicoureteral reflux (VUR).

Introduction

Vesicoureteral reflux is a polygenic genetic disorder with an incidence of 1% in the general population.¹ Twinning studies demonstrate a 100% concordance in identical twins and 35-50% prevalence in fraternal twins when tested early in life.² Cystography of siblings and offspring of patients with VUR has shown a high prevalence of VUR.³ Screening of offspring and siblings has been proposed as a means to detect a population at risk and allow timely treatment in order to reduce the risk of adverse outcomes associated with VUR, including urinary tract infections (UTI), pyelonephritis and renal scarring. In the 1997 VUR guideline,⁴ a proposed future research goal was to determine the impact of sibling and offspring screening with early medical or surgical treatment on the risk of these outcomes.

Methodology

Literature Search, Data Extraction, and Evidence Combination

A meta-analysis of the existing literature was performed to develop recommendations for sibling and offspring screening. Outcomes included the VUR prevalence among siblings and offspring of VUR index cases who have been screened by cystography, and the prevalence of renal damage as determined by intravenous pyelogram (IVP) and technetium-99m-labeled dimercaptosuccinic acid (DMSA) scanning. The potential differences in the estimates of prevalence rates were stratified by the patient characteristics of age, sex, VUR grade, renal scarring and symptoms. Where possible, the effects of treatment were assessed in regards to resolution, infection and renal scarring.

Data from 22 articles (published between 1975 and 2006) were extracted and meta-analyzed. These reports included 3,201 children (2,957 siblings and 244 offspring); 3,040 children were screened with a cystogram (2,796 siblings and 244 offspring).

Outcomes Analysis

VUR Prevalence among Siblings and Offspring

Estimates of VUR prevalence among siblings ranged between 2.9% and 51.5%, for a mean estimate of 27.4 per 100 siblings screened (95% confidence interval [CI]: 19.6, 36.9). The VUR prevalence among mixed asymptomatic and symptomatic siblings was 27.7 per 100 (95% CI: 16.8, 42.2) siblings screened and 28.5 per 100 (95% CI 18.2, 41.7) asymptomatic siblings screened. For offspring, the prevalence of VUR ranged between 21.2% and 61.4%. The overall prevalence of VUR among the four studies was 35.7 per 100 (95% CI: 16.4, 61.0) offspring screened.

The influence of age at the time of screening was assessed to attempt to identify the causes of the wide variations in reported prevalence. The prevalence rate based on age at screening was estimated from the equation: Prevalence (per 100 screened) = 53.8 -0.3 x age (mo). The prevalence of VUR in siblings decreased with increasing age at screening by a rate of one screened person every 3 months (Figure 1), suggesting a higher prevalence in the youngest screened sibling (please refer to the Chapter 4 Technical Report for the derivation of this estimate).

Figure 1. VUR prevalence rate among siblings by average age at screening (open circles=asymptomatic; filled squares=mixed symptomatic and asymptomatic)

The VUR prevalence in siblings when stratified by sex is based on eight studies. The prevalence in screened male patients (21.3%) was not statistically different from that of females (26.4%); therefore, the sex of the sibling is not a selection criterion for screening.

VUR grade was assessed in 68% of the screened siblings. The overall prevalence in 11 studies that recorded VUR grade was 29.6 per 100 screened siblings. The prevalence for VUR grades I– II was 16.7 per 100 screened siblings and the prevalence for VUR for grades III–V was 9.8 per 100 screened siblings; these rates were not significantly different. Finally, laterality was assessed in screened siblings. Of those with VUR, there were similar percentages of unilateral and bilateral reflux (17.1 vs. 15.1 per 100 screened siblings, respectively).

Renal Cortical Abnormalities

Nine sibling studies provided information on both VUR and renal cortical abnormalities as diagnosed by either DMSA scanning or intravenous pyelogram (IVP). The prevalence of renal cortical abnormalities ranged between 11% and 54% for an overall estimate of 19.3 per 100 (95% CI: 10.9, 32.0) siblings with reflux. Among asymptomatic siblings, the prevalence of renal cortical abnormalities ranged between 0% and 100% for an overall estimate of 14.5% (95% CI: 7.2, 27.3) and in studies where both symptomatic and asymptomatic siblings were examined, the overall estimate was 22.8% (95% CI: 7.2, 53.1). When evaluating only those studies in which 90% or more of the patients with VUR, or those initially screened for VUR, underwent DMSA scanning, the prevalence rate for renal scarring was 18.8% (95% CI: 9.8, 33.2). There was only one study that directly assessed the prevalence of renal cortical abnormalities among siblings with or without a prior history of UTI.5 In this study the prevalence of renal cortical abnormalities was significantly different, with a rate of 35.2% (95% CI: 24.2, 47.8) among those with a prior history of UTI and 11.7% (95% CI: 6.2, 21.0) (p < 0.05) among those without a prior UTI.

There was limited information with which to assess the association between VUR grade and renal cortical abnormalities. Of the nine studies providing both rates of VUR and renal cortical abnormalities, only two assessed cortical abnormalities by VUR severity.^{6, 7}

In the 132 children (out of 144 total cases of VUR) who underwent a renal scan, 18 cases of scarring were found. Seven had grade II VUR, eight had grade III, and three had grade IV. The best association that can be estimated with the available data is at the aggregate level. Numerically, the association between the prevalence of VUR and the prevalence of renal cortical abnormalities by DMSA/IVP was moderate (Pearson r = 0.45, p = 0.32). However, the association across the seven studies providing information on both severity of VUR and renal cortical abnormalities was weak.

Recommendation: In siblings of children with VUR, a voiding cystourethrogram is recommended if there is evidence of renal scarring on ultrasound or if there is a history of urinary tract infection in the sibling who has not been tested.

[Based on Panel consensus]

Option: Given that the value of identifying and treating VUR is unproven, an observational approach without screening for VUR may be taken for siblings of children with VUR, with prompt treatment of any acute urinary tract infection and subsequent evaluation for VUR.

[Based on Panel consensus]

Option: Sibling screening of older children who are toilet trained may be offered, although the value of identification of VUR is undefined.

[Based on Panel consensus]

Option: Ultrasound screening of the kidneys in the sibling of a child with VUR may be performed to identify significant renal scarring, and to focus attention on the presence and potential further risk of VUR.

Option: Screening offspring of patients with VUR can be considered as similar to screening of siblings.

[Based on Panel consensus]

Summary

VUR was detected at an overall rate of 27 and 36 per 100 children screened among siblings and offspring, respectively, which is significantly greater than the 1% estimate in the general population.¹ There were no data to evaluate the immediate and long-term effects of treatment in the screened population of siblings. Consequently these recommendations are based on the assessment of present management and treatment, evaluation of the literature, and the results of this meta-analysis.

There were insufficient data to evaluate whether siblings with VUR are at greater risk of developing UTI than the general population or the rate of resolution in siblings with VUR. One report suggested that the grade-specific time to VUR resolution was shorter in asymptomatic siblings compared with symptomatic children.⁸

It is recognized that children with UTI and VUR are at increased risk for pyelonephritis. In this meta-analysis (see Chapter 1) children with VUR were found to be 2.8 times more likely to develop a renal scar (3.7 times more likely for the refluxing renal units) than children with pyelonephritis and no reflux. Only one study directly assessed the prevalence of renal damage and VUR among children with or without UTI at screening. In this study the prevalence of renal damage was significantly greater among those with a prior history of UTI and older age.⁵

In this Guideline, "congenital" renal cortical abnormalities were present in 19% of those with prenatal hydronephrosis without UTI. The rate of renal damage in infants with VUR diagnosed without UTI is 14%. In contrast, the rate of renal damage in screened siblings, some of whom were symptomatic, was 22.8%; this suggests that renal damage may be preventable in some cases. With treatment, either medical or surgical, new renal scarring has been estimated to occur in 5.2% to 31.4% of patients.⁴ The rate of new scarring in medically observed, untreated patients with VUR is under study but presently unknown; therefore, CAP treatment of the youngest (<1 year of age) screened sibling with VUR seems prudent at this time.

Parental preferences need to be considered when screening siblings and offspring. Parents with personal experience with VUR, VUR nephropathy, renal insufficiency, and hypertension will have specific opinions for their children. If one of their children has experienced a febrile UTI or pyelonephritis and is diagnosed with VUR this may influence their choice as to screening of offspring and siblings. Guidance as to how to perform screening and in whom, will be influenced by physician recommendations.

Due to the insufficiency of published data, the value of sibling screening for VUR cannot be determined; therefore, the following recommendations for improving the study design are proposed. All siblings must have an initial DMSA renal scan and a subsequent scan in the event of a documented UTI; the time to spontaneous resolution must be accurately recorded. The age at screening, the distinction between symptomatic and asymptomatic siblings, the presence and type of urinary infections, the duration of VUR, VUR grade and corresponding renal damage, if present, should be documented. Documentation of renal damage in all siblings at screening regardless of VUR status, if possible, is necessary. Genetic assessment of proband and siblings to determine the relative risk for UTI, VUR and renal scarring should be performed.

References

1. Arant, B. S., Jr.: Vesicoureteric reflux and renal injury. Am J Kidney Dis 1991; 17: 491.

2. Kaefer, M., Curran, M., Treves, S. T. et al.: Sibling vesicoureteral reflux in multiple gestation births. Pediatrics 2000; 105: 800.

3. Jerkins, G. R., Noe, H. N.: Familial vesicoureteral reflux: a prospective study. J Urol 1982; 128: 774.

4. Elder, J. S., Peters, C. A., Arant, B. S., Jr. et al.: Pediatric Vesicoureteral Reflux Guidelines Panel summary report on the management of primary vesicoureteral reflux in children. J Urol 1997; 157: 1846.

5. Pirker, M. E., Colhoun, E., Puri, P.: Renal scarring in familial vesicoureteral reflux: is prevention possible? J Urol 2006; 176: 1842.

6. Celik, A., Ulman, I., Aydin, M. et al.: Familial vesicoureteral reflux in asymptomatic siblings. Turk J Pediatr 2002; 44: 240.

7. Wan, J., Greenfield, S. P., Ng, M. et al.: Sibling reflux: a dual center retrospective study. J Urol 1996; 156: 677.

8. Parekh, D. J., Pope, J. C., 4th, A., M. C. et al.: Outcome of sibling vesicoureteral reflux. J Urol 2002; 167: 283.

Unabridged Version of this Guideline [pdf]

Index patient

The healthy neonate with unilateral mild (Society for Fetal Urology [SFU] grade 1) to moderate (SFU grades 2–3) hydronephrosis identified on a screening prenatal ultrasound at 30 weeks gestation.

Introduction

Postnatal evaluation of prenatal hydronephrosis (PNH) offers the opportunity to diagnose and manage vesicoureteral reflux (VUR) before secondary damage occurs from urinary tract infection (UTI). Prior studies in patients with ipsilateral renal function loss but no history of infection have already led to the recognition that maldevelopment in the absence of infection can contribute to the clinical picture of "reflux nephropathy". Evaluation of affected infants could additionally provide insights into the benefits, if any, of early identification of VUR and institution of continuous antibiotic prophylaxis to prevent UTI, pyelonephritis and renal scarring.

Methodology

Literature Search, Data Extraction, and Evidence Combination

A meta-analysis of the existing literature was performed to determine the impact of postnatal management of neonates/infants with PNH in prevention of UTI, pyelonephritis, and new renal scarring, and to develop recommendations for screening to detect VUR. Outcomes included the incidence of VUR (stratified by sex, laterality, and postnatal diagnosis of reflux), incidence of UTI, and incidence of renal cortical abnormalities. Of the 43 studies selected for meta-analysis involving at least 6,579 infants with PNH (who were evaluated postnatally), there were 34 studies reporting VUR incidence among 4,756 infants with PNH, 12 studies providing information on renal abnormalities in 530 infants, and eight studies reporting UTI in 616 infants. There was large variability across these studies regarding the definition of PNH, indications and timing of postnatal evaluation; therefore the percentage of patients undergoing renal ultrasonography, cystography, or other investigations also varied. Articles that included other diagnoses, (such as posterior urethral valves, ureterocele, etc.) that could not be separated were excluded from analysis.

Outcomes Analysis

Incidence of Vesicoureteral Reflux

The incidence of VUR in patients with PNH was reported by 34 studies. Of these, the mean percentage of neonates/infants screened by cystography was 78% (range, 11% to 100%). Reflux was detected in 7% to 35% of patients undergoing cystography, averaging 16.2% (Figure 1).

Figure 1. Forest plot of VUR incidence rate among infants with PNH

Reflux per renal unit with PNH was determined from 15 studies, yielding a mean of 12.6% (95% CI: 8, 18%) as shown in Figure 2.

Figure 2. Forest plot of reflux incidence rate among renal units with PNH

Some studies did not perform cystography if postnatal renal ultrasonography was normal, potentially leading to an erroneous estimate of VUR incidence. Considering the 16 studies in which 100% of neonates/infants with PNH and hydronephrosis confirmed postnatally underwent cystography, the mean percentage of patients with VUR was 15.2% (95% CI: 10.9, 20.7). Considering only those studies in which 100% of patients with PNH underwent cystography (even if no hydronephrosis is detected postnatally), the incidence of reflux was 18.2 (95% CI: 13.0, 24.8) per 100 infants.

In eight studies, information was provided regarding VUR into a contralateral normal kidney. VUR into the nondilated kidney accounted for a mean of 25.2% (95% CI: 17.6, 34.7) of detected reflux. Considering the total of 3,082 renal units screened, there was a mean incidence of 4.1% VUR (95% CI: 2.3, 7.4) into a nondilated kidney.

Table 1 provides estimates for VUR incidence stratified by sex, laterality, and postnatal diagnosis of reflux. The distribution of VUR grade was approximately one-third grades I–II, one-third grade III and one-third grades IV-V based upon maximum grade in both patients and renal units. Approximately one-half of the patients had bilateral VUR.

Females with a prenatal diagnosis of PNH had a significantly higher (p = 0.022), incidence of VUR compared to male infants. In nine studies the mean incidence of reflux was 17.0% (95% CI: 10.7, 25.8) (17.0% vs. 15.6%, p =0.59), in children with PNH having a normal postnatal ultrasound. Therefore the incidence of VUR in neonates/infants with a history of PNH was the same regardless of postnatal renal ultrasonography findings (Table 1).

Table 1. VUR incidence rate per 100 infants screened

*Includes only studies in which VCUG was performed.

RPD, renal pelvic diameter; US, ultrasound; VCUG, voiding cystourethrogram

Note: N corresponds to the number of neonates/infants screened. Numbers within parenthesis correspond to the 95% confidence interval.

PNH is determined by the anterior-posterior renal pelvic diameter (RPD) measured in transverse section. However, no universally accepted threshold defines patients who are most likely to benefit from postnatal evaluations. Figure 3 shows the incidence of VUR based upon the minimum prenatal and postnatal RPD used to prompt screening cystography. Increasing RPD did not predict an increased likelihood of VUR, with RPD of only 4 mm associated with reflux in approximately 10% to20% of neonates/infants screened.

Figure 3. VUR incidence by prenatal and postnatal RPD thresholds (Open triangles=3rd trimester as first evaluation; Open circles=outliers; Filled squares=all other

Other factors that might influence the incidence of VUR were investigated, including trimester of PNH assessment, timing of postnatal evaluation (from 1 to 3 months), and percentage of patients screened. None of these factors had a correlation with findings of reflux.

Incidence of Urinary Tract Infections during Screening Period

Eight studies reported data on the occurrence of UTI during the postnatal screening period. There was considerable variability in antibiotic administration, with some instituting prophylaxis in all patients after delivery, while others only prescribed antibiotics to those found to have VUR and/or PNH. Agents used were not uniformly reported. Consequently the risk of UTI in patients with PNH with and without VUR cannot be determined, nor can the possible impact of antibiotic prophylaxis. Within the variable conditions mentioned above, the incidence of reported UTI ranged from 0.5–21.3 cases per 100 infants under surveillance, averaging 4.2%; most of these cases were in patients with VUR.

Renal Cortical Abnormalities

Of 43 extracted publications, five reported DMSA screening in a mean of 60 (range 23–155) patients, while nine reported a mean of 79 (range 24–236) renal units screened with DMSA nuclear scintigraphy. There was heterogeneity in descriptions of these renal abnormalities, variously categorized as focal or diffuse renal cortical scars and/or decreased global uptake. Of studies reviewed, 94% of patients and 98% of renal units with VUR underwent DMSA screening in the early postnatal period. The incidence of any renal abnormality before UTI in patients with VUR ranged from 2–63% of patients with a mean of 21.8%. The incidence of abnormalities per renal unit ranged from 26% to 42%, with a mean of 32.3%.

In seven of nine studies reporting abnormalities by renal unit, reflux grade was also available. For this meta-analysis, VUR severity was categorized as low-moderate (grades I-III) and severe (grades IV-V). The overall prevalence of renal abnormalities was 29.4 per renal unit (Figure 4).

In the low-moderate group (grades I–III) the prevalence was 6.2% versus 47.9% in those with severe reflux (grade IV–V) (p<0.0001).

Figure 4. Forest plots of renal cortical abnormality rates by reflux grade

Recommendation: Voiding cystourethrogram is recommended for children with high-grade (Society of Fetal Urology grade 3 and 4) hydronephrosis, hydroureter or an abnormal bladder on ultrasound (late term prenatal or postnatal), or who develop a urinary tract infection on observation.

[Based on review of the data and Panel consensus]

Option: An observational approach without screening for VUR, with prompt treatment of any urinary tract infection, may be taken for children with prenatally detected hydronephrosis (SFU grade 1 or 2), given the unproven value of identifying and treating VUR. It is also considered an option to perform a voiding cystourethrogram in these patients to screen for VUR.

[Based on Panel consensus]

Summary

Within the limitations mentioned above, this meta-analysis found that approximately 16% of neonates and infants with PNH have reflux. This incidence was independent of prenatal RPD, with reflux also detected in non-dilated renal units. The incidence of VUR was similar when postnatal urinary tract ultrasonography was normal. Reflux grade was III or greater in two-thirds of patients, with renal abnormalities occurring in nearly 50% of those with grades IV–V.

Given the widely accepted view that VUR occurs in only approximately 1% of otherwise normal infants,¹ those with PNH represent a group at increased risk for reflux. Furthermore, the reflux grade was more severe than expected in older children presenting with UTI. The finding that renal abnormalities are already present in a significant percentage of renal units with higher grades of reflux raises concern for the additional negative impact even a single UTI might impart. Together these considerations potentially support postnatal cystography in all neonates with PNH, an age group in which cystography also may be better accepted by families. However, the lack of a prospective study demonstrating benefit of reflux detection in asymptomatic neonates and recent data from prospective studies that question efficacy of antibiotic prophylaxis to prevent UTI make screening cystography an option, rather than a recommendation.

References

1. Arant, B.S., Jr.: Vesicoureteric reflux and renal injury. Am J Kidney Dis 1991; 17: 491.