Managing vesicoureteral reflux in children: making sense of all the data

Defining urinary tract infection

In the majority of cases in children, VUR is diagnosed after a febrile UTI episode or abnormality seen on ultrasound imaging, potentially more common in the prenatal hydronephrosis population, leading to a VCUG being conducted. Attention to detail on how a urine specimen was obtained and whether pyuria is present is key in the diagnosis of a UTI. Accurate diagnosis of UTI is a critical turning point in the overall algorithm for how the patient will be evaluated and guides major clinical decisions. These decisions often hinge on the presence of recurrent infection and may lead to repeat imaging or medical/surgical intervention.

In 2011, the AAP revised the guidelines for the diagnosis and management of initial UTI in febrile infants from 2 to 24 months. The guidelines stress the importance of appropriate urine specimen collection, and a diagnosis of UTI is made only when at least 50,000 colonies per milliliter of a single uropathogenic species is present along with pyuria. The specimens should be collected via catheterization or suprapubic aspirate and a urine culture sent. If the patient has a febrile UTI, he or she will receive a renal and bladder ultrasound; if there are no anatomical abnormalities, no further imaging is conducted at that time⁴. Midstream urine collection from toilet-trained children is acceptable but should be carried out in males with foreskin retracted. These guidelines have led to a change in the threshold to try to make a diagnosis of VUR and VCUG is not automatically part of the initial febrile UTI workup unless anatomical abnormalities are seen on the screening ultrasound. This has led to a decrease in the number of VCUGs performed and a steady decline in procedures performed for VUR⁵. To date, however, it does not appear to have enriched the yield of significant reflux in the populations being tested⁶.

Imaging evaluation

The decision to perform voiding cystography, which is viewed as an invasive exam because of the need for catheterization, should be appropriately weighed with patient history, including BBD, recurrent infections, changes seen on renal ultrasound, and patient and parental desires. VCUG remains the gold standard, however, in terms of defining the presence, grade, and possible impact of VUR (Figure 1). Ongoing attempts to standardize this study for modern usage are underway⁷. The principal techniques used in the evaluation of the renal impact of VUR include dimercaptosuccinic acid (DMSA) scanning (Figure 2) and renal ultrasound because of their lack of need for sedation or catheterization. However, DMSA scans may continue to become more limited because of accessibility and radiation exposure to the patient. Veenboer et al. showed that, out of 242 adult renal units evaluated with spinal dysraphism, DMSA scintigraphy could demonstrate more renal scars than ultrasound at a rate of 45.9% compared with 10.3% of renal units, demonstrating the greater sensitivity of DMSA scanning for renal scarring⁸. The access to the tracer technetium-99m DMSA and need for peripheral intravenous placement have limited the use of renal scans in certain parts of the world in recent years. Ultrasound remains the most feasible upper tract imaging modality because it is the least invasive and most readily available. The ability to detect large cortical defects and renal asymmetry or discrepancies in renal growth makes it a reasonable initial and follow-up imaging modality.

Figure 1. Voiding cystourethrogram showing bilateral grade 3 vesicoureteral reflux with a large, smooth-walled bladder.

The calyceal configuration is normal without evidence of clubbing, which might have suggested renal parenchymal scarring.

This is an original, unpublished image obtained by the author for this publication.

Figure 2. Dimercaptosuccinic acid scan showing significant right renal scarring and more focal cortical abnormalities on the left.

This is an original, unpublished image obtained by the author for this publication.

This approach of imaging the upper tract prior to VCUG was popularized as the “top-down approach” (TDA). TDA operates under the assumption that clinically significant VUR will have associated changes on acute DMSA renal scans and are therefore worthwhile to detect but assumes that in patients with normal DMSA scans any VUR is clinically insignificant. With this approach, a patient receives a DMSA scan shortly after a UTI episode and if abnormalities are present this is followed up with VCUG. Preda et al. prospectively evaluated the TDA in 290 patients younger than 1 year of age who presented with UTI (79% febrile infections)⁹. This revealed that 51% of patients had an abnormal DMSA scan; of those, 85% (44 out of 52) were found to have VUR. Of the remaining patients, 8 were found to have VUR with a normal DMSA: low-grade VUR was diagnosed in 7 and grade 3 VUR in 1 male⁹. This has been countered by a Cochrane Review conducted in 2016 by Shaikh et al., who concluded that DMSA lacks the accuracy to detect VUR of any grade and should be challenged as a screening test because of the high proportion of children labeled at-risk for high-grade VUR¹⁰. Interestingly, this study found that children with a negative DMSA scan have a less than 1% probability of having high-grade VUR¹⁰.

Numerous other novel markers—such as delta neutrophil index, neutrophil-to-lymphocyte ratio, D-dimer, erythrocyte sedimentation rate, procalcitonin, and C-reactive protein—have been introduced as potential triggers for imaging when evaluating patients with UTI. Shaikh et al. conducted a randomized controlled trial of 309 children with their first febrile UTI from age 1 to 24 months who underwent either a biomarker screening as the threshold or DMSA changes to trigger a VCUG¹¹. It was shown that top-down and biomarker-based screening for VUR had poor sensitivity, as they did not detect 33% and 29% of high-grade VUR, respectively¹¹.

Evolution of vesicoureteral reflux management

The overarching goal of all VUR management is to prevent UTI and renal scarring by using the least invasive means to preserve maximum renal function and prevent hypertension. This becomes very multifactorial once you add parental and practitioner bias along with the length of time the patient has been diagnosed and treated. VUR management has evolved to include medical management or observation as the first choice for most reflux patients, and the approach must be individualized to focus on preventing UTI instead of only VUR resolution. This is demonstrated by the high likelihood of spontaneous resolution of reflux with correction of BBD, which decreases infections. It is broadly accepted, with some caveats, that sterile reflux does not pose any real harm to the kidneys. All treatment modalities should be considered when developing an individualized treatment plan based on patients’ risk stratification, compliance of the family, and access to care among other considerations. The overall clinical picture may lead providers to offer definitive correction in certain patients over medical management or active surveillance.

In consideration of the AAP guidelines, patients may not be diagnosed until after two pyelonephritic episodes or may be diagnosed at a more advanced age with renal scarring already present. In these patients, providers may be more inclined to encourage definitive treatment, rather than observation or continuous antibiotic prophylaxis (CAP) prior to another pyelonephritic episode. On the other end of the spectrum, good results have been shown by providers who are starting to wean off CAP therapy in toilet-trained children with resolution of BBD combined with close observation for the development of constipation or infections¹². The idea of active surveillance or cessation of CAP reinforces the need to prospectively risk-stratify the patients who are at risk for breakthrough UTI on CAP. Hidas et al. showed that patients who presented after a UTI episode, were female, or had high-grade (4 and 5) VUR are the most concerning populations¹³. Identifying those patients at risk for recurrent UTI when taken off prophylaxis is an even more important group, as they pose a real and challenging risk.

International reflux study in children

This study included children younger than 11 years old with high-grade VUR (grade 3 or 4) who presented with UTI and were randomly assigned to CAP or corrective open surgery with an American²² and a European²³ arm. The European arm also recorded the presence of BBD in the patient population (18%), and the BBD, when untreated, led to more UTI and longer persistence of VUR along with grade variation during follow-up²⁴. Some of the surgically managed patients had complications with post-operative obstruction^25,26 but overall had a decreased rate of pyelonephritis compared with CAP, and there was an equal incidence of UTI²³ in both groups and equal efficacy in reducing new renal scarring. At the 5-year follow-up, there was no significant difference between the surgically and medically managed groups in the development of new renal scarring or scar progression²⁷, but there were more new scars observed in patients who entered the study at a younger age or had parenchymal thinning present²⁸. This study therefore demonstrated the equivalence of medical and surgical therapy; however, the incidence of surgical obstruction reported in the European arm was higher than in the American arm and greater than in many other clinical reports.

Swedish study

This trial is a multi-center prospective trial of 203 children (128 females and 75 males) from age 1 to 2 with dilating VUR grade 3 or 4 who received a VCUG and renal scan at entry in the study and then were randomly assigned to CAP (69), endoscopic treatment (66), or surveillance (68)²⁹. At 2 years of follow-up, all treatment arms showed reflux improvement with downgrading to grade 1 or 2 or resolution: prophylaxis group (39%), endoscopic group (71%), and surveillance (47%) improvement³⁰. There were 67 recurrent febrile UTIs in 42 females and 7 males. Females with recurrent febrile infections occurred in 19% of prophylaxis, 23% of endoscopic, and 57% of surveillance groups.

The authors concluded that females with recurrence of infection had a higher tendency to have persistent VUR after 2 years, but reflux grade at induction of study (whether grade 3 or 4) was not predictive of recurrence³¹. A strong association with new renal damage and recurrent UTI was seen in the female population, all of whom were in the surveillance and endoscopically managed patient groups³². The importance of this study is the inclusion of a surveillance cohort of patients along with an increased male patient population. BBD was demonstrated by urodynamic evaluation and was linked to recurrent UTI and reflux persistence as well as new scarring.

PRIVENT trial

In Australia, a placebo-controlled trial with trimethoprim-sulfamethoxazole (TMP-SMX) was conducted with CAP for children younger than 18 years of age (median age was 14 months and 53% had grade 3 VUR or higher). Not all patients had VUR. A 6% decrease in UTI occurrence was seen in the CAP group over the placebo, and the Bactrim-receiving patients had a UTI rate of 13% (36 out of 288) and the placebo group a rate of 19% (55 out of 288)³³. The authors concluded that CAP has a limited role, but the main criticism during this study was that no upper tract imaging was conducted to be able to comment on the presence of renal scarring between the two groups.

RIVUR trial

RIVUR was a multi-center, double-blinded, randomized, placebo-controlled trial that evaluated 607 children (558 girls and 49 boys) who ranged in age from 2 to 71 months and who had grade 1 (11%), grade 2 (42%), grade 3 (38%), and grade 4 (8%) (grade 5 excluded) VUR that was diagnosed after UTI episode^34,35. Patients were randomly assigned to placebo versus TMP-SMX antibiotic prophylaxis; this is very similar to the PRIVENT trial except that patients received DMSA at baseline and after 1 and 2 years of follow-up. After 2 years of follow-up, the CAP group had a 50% lower risk of UTI recurrence. However, 91.9% of the study population was female and VUR grade 1 to 3 was present in 91.7% of patients³; therefore, the clinical significance has been challenged because of the attributes of the patient population, as it represents a relatively low-risk population. The most significant benefit with CAP was seen in those subgroups who had BBD or presented with a febrile UTI. DMSA scans at follow-up did not show any difference in new renal scarring between the two patient populations, although reduction in scarring was a secondary endpoint and the study was not powered to fully assess this feature. Although the reflux grading system used in the RIVUR trial has been used for decades, there was still relatively low inter-observer agreement on reflux grade. In the context of the study, three experts adjudicated grade, yet this observation would suggest that the use of reflux grade as a risk-stratifying factor may have significant clinical limitations³⁶. Recent investigations have suggested that using distal ureteral diameter as a “grading” scale for reflux severity may be more reliable than the traditional system^37,38.

Risk stratification

With regard to the future of VUR management, the best compromise is to identify the patients with the greatest risk of VUR sequelae and potentially never discover incidental VUR in a low-risk patient. A compelling concept is reclassifying the data from prior randomized controlled trials using risk classification and seeing whether the clinically significant risk factors can be identified with an acceptable level of morbidity. In 2018, the data from the RIVUR trial were re-evaluated using risk stratification³⁹. The clinical characteristics used to develop the risk stratification criteria are based on the model by Hidas et al. on UTI recurrence¹³. Low-risk patients were defined as circumcised males with no BBD present and with grade 1 to 3 VUR. High-risk patients were classified as uncircumcised males with a grade 4 VUR or as females with BBD.

There were 385 (63.9%) patients in the low-risk population and 217 (36.1%) patients in the high-risk population for the RIVUR trial. Wang et al. showed that there is no significant difference in UTI recurrence in the low-risk group between CAP and placebo (p = 0.151) but a higher UTI recurrence in the high-risk population and 31.5% reoccurrence in placebo versus 11.4% in CAP patients (p = 0.001)³⁹. Patients in the high-risk cohort receiving placebo had a 3.7-fold increased risk of UTI recurrence over the 2-year period when compared with their CAP high-risk cohort, but no difference was found in renal scarring in the high- or low-risk grouping.

That no difference was seen in new or worsening renal scarring between the populations of this trial could be rooted in the fact that the majority of patients were female (91.9%) and had VUR grade 1 to 3 (91.7%)³; therefore, the main characteristic affecting their risk stratification is BBD. To broadly apply the concept of standardized risk stratification to a more diverse population by combining the findings of the RIVUR and Swedish trials may lead to a more concrete understanding of risk stratification and the new frontier of VUR management.

Bladder and bowel dysfunction

During the RIVUR trial, the dysfunctional voiding scoring system⁴⁰ was used to characterize patients. In the reclassification of the RIVUR data, patients who were not toilet-trained and had fewer than one bowel movement per day were categorized as having BBD. With these criteria, it was found during the reclassification that being a non-toilet-trained child with BBD was an independent predictor of breakthrough UTI in the CAP group³⁹.

The importance of the management of BBD is emphasized in the 2010 meta-analysis conducted by the American Urological Association, which showed that BBD was associated with higher UTI incidence while on antibiotic prophylaxis, lower rates of resolution of VUR, and reduced success of endoscopic treatment of VUR⁴¹ (Figure 3). Children with VUR and coexisting BBD who underwent intervention with timed voiding, biofeedback, anticholinergic administration, or alpha blockers to address the lower urinary tract symptoms or voiding pattern were found to downgrade their reflux on average by two or more grades¹⁵. In a study of 30 children with VUR, downgrading or resolution of VUR was seen in 70% of patients when treatment for BBD was incorporated into their treatment plan in the form of anticholinergic medication⁴².

Figure 3. Voiding cystourethrogram showing left reflux as well as a dilated urethra in a girl with bladder and bowel dysfunction.

This is the so-called “spinning top” urethra that results from incomplete relaxation of the external sphincter/pelvic floor.

This is an original, unpublished image obtained by the author for this publication.

The incorporation of diagnosis and treatment of dysfunctional voiding and constipation into the initial evaluation of the patient could continue to improve the overall outcomes and shape areas of future research. Shaikh et al. were able to combine data from RIVUR and CUTIE (Careful Urinary Tract Infection Evaluation), which is a population of children younger than 6 years old who have had a UTI and were followed for 2 years⁴³. In this population, 54% of the 181 toilet-trained children had BBD (daytime incontinence, holding maneuvers, or constipation). Results showed a higher risk of developing UTI if VUR and BBD were both present when compared with children with VUR alone or BBD alone. Children with VUR and BBD were shown to have the greatest benefit from antibiotic prophylaxis⁴³.

Endoscopic management

An option popularized over the last two decades is cystoscopic subureteric injection, which has been accomplished with several different materials over the years but currently is performed mostly with dextranomer/hyaluronic acid microspheres (Deflux). The goal of subureteric injection is to instill a bulking agent in the submucosal intramural tunnel in hopes of downgrading or resolving VUR. The endoscopic management has migrated from the subureteric Teflon injection (STING) technique (which originally used polytetrafluoroethylene injected below the ureteric orifice at the 6 o’clock position to generate a “crescent shaped” orifice)⁴⁴ to HIT (hydrodistention of the ureteric orifice and the injection of bulking agent is located in the mid to distal submucosal tunnel at the 6 o’clock position)⁴⁵ and now more commonly uses the double HIT technique (modified HIT technique with proximal and distal intraluminal submucosal injections)⁴⁶. A meta-analysis conducted by Yap et al. showed that the overall resolution of VUR was 82.5% with the HIT technique compared with 71.4% with the STING technique (p <0.00001)⁴⁷. There are numerous publications on the success rate of VUR resolution by VUR grade, and the highest attainment was seen in lower grades of VUR^48,49.

This technique, though an attractive option to avoid prolonged antibiotic exposure and invasive surgical treatments, may be limited by its durability over time. During the Swedish trial, the endoscopic cohort had recurrent dilating reflux in 20% of patients after 2 years as well as others with recurrence to grade 1 or 2 VUR³⁰. It should be noted that the STING procedure was used in the Swedish trial and initial success was very low. Currently reported success rates (cure of VUR or prevention of febrile UTI) using the double HIT method exceed 90%, and only 5% went on to open surgery after a single endoscopic treatment⁵⁰. Some may feel that the benefit of Deflux is to spare the patient from antibiotic prophylaxis while waiting for the natural resolution of VUR or coordinated maturation of voiding with toilet training to occur, even though its longevity over time is potentially inferior. The counterargument to this rationale is that the procedure requires general anesthesia and can take multiple injections, which come with their own concerns for patients.

National trends in VUR management were studied in 2014 by Herbst et al., who looked at 14,430 patients (17,826 procedures), of whom 49% underwent reimplantation and 51% received injection with dextranomer/hyaluronic acid from 2004 to 2011 in the PHIS (Pediatric Health Information System) database⁵¹. The total number of VUR procedures has steadily declined through this time interval, but the average number of reimplantation surgeries remained steady, leading the author to conclude that this trend is declining for dextranomer/hyaluronic acid injections as a VUR treatment. However, in a similar study by Garcia-Roig et al.⁵, the PHIS database was further evaluated to 2015, several years after publication of the 2011 AAP UTI guidelines. A total of 43,431 VCUG encounters and 28,484 anti-reflux procedures were evaluated (57% reimplantation, 41% dextranomer/hyaluronic acid injection, and 2% laparoscopic ureteral reimplantation). Following publication of the UTI guidelines, a significant overall drop in surgical intervention was noted for all procedure types. Although robotic surgical intervention comprised only 2% of all procedures, its utilization actually rose significantly after 2011⁵.

Minimally invasive surgery

In the modern setting, patients and their families desire the least invasive option with the highest success rate and lowest complication rate; this lends itself to the current interest in RALUR. The introduction of RALUR in 2004⁵² has led multiple centers to adopt the technique, and most are able to achieve favorable outcomes approaching that of the gold-standard open-surgical approach^53,54. There is a known steep learning curve to robotic surgery⁵⁵, and mastering the technique to match the reported outcomes is not universally experienced by all surgeons^56,57.

The inherent benefits of robotic surgery are improved visualization and magnification, ergonomic comfort for the surgeon, cosmetic incisions⁵⁸, stabilization of movement with wristed instruments, shorter hospital stay⁵⁹, and decreased narcotic requirements⁶⁰. The downside is the increased cost⁶¹ of robotic surgery and access to a robotic system along with a pediatric dedicated robot team⁶².

With regard to current trends in VUR management, RALUR may not quickly replace the gold-standard open-surgical approach but is currently a viable option for patients presenting to surgeons with a robotic skillset and a dedicated robotics program. Some providers contend that RALUR has the most promise in caring for patients who are older or have complex anatomy⁶³. The future benefit of the robotic approach may lie in defining the patient characteristics that lead to a successful outcome⁶⁴ and may be able to be tied into the pre-operative risk stratification or anatomical considerations.