- Email: [email protected]
Most Infants with Dilating Vesicoureteral Reflux can be Treated Nonoperatively
Most Infants with Dilating Vesicoureteral Reflux can be Treated Nonoperatively Aaron D. Martin,* Muhammad W. Iqbal, Bruce M. Sprague, Mireya Diaz, H. Gil Rushton, Craig A. Peters, Massoud Majd and Hans G. Pohl From the Children’s National Medical Center, Washington, D.C., and Vattikuti Urology Institute, Henry Ford Hospital (MD), Detroit, Michigan
Abbreviations and Acronyms APN ¼ acute pyelonephritis bUTI ¼ breakthrough febrile UTI DMSA ¼ dimercapto-succinic acid UTI ¼ urinary tract infection VCUG ¼ voiding cystourethrogram VUR ¼ vesicoureteral reflux Accepted for publication August 26, 2013. Study received Children’s National Medical Center institutional review board approval. * Correspondence: Louisiana State University Health Sciences Center and Children’s Hospital New Orleans, Department of Urology, 200 Henry Clay Ave., New Orleans, Louisiana 70118 (telephone: 504-896-9233; FAX: 504-896-9861; e-mail: [email protected]).
Purpose: Dilating vesicoureteral reflux provokes concern for physicians and parents that often leads to corrective surgery in young children. Since there are limited data describing the natural history of dilating vesicoureteral reflux in infants, we identified factors predictive of resolution/improvement in infants initially treated nonoperatively. Materials and Methods: We reviewed the medical records of 90 infants 6 months old or younger from 2004 to 2010 who were referred for prenatal hydronephrosis or initial febrile urinary tract infection and found to have dilating vesicoureteral reflux (grade 3 or greater). Variables of interest included presentation, dimercapto-succinic acid results, sex, breakthrough febrile urinary tract infections, reflux grade and bilateral reflux. Cox regression analysis was performed to determine predictors of spontaneous resolution and/or improvement to reflux grade less than 3 as well as predictors of surgical intervention. Results: Included in final analysis were 80 infants (113 renal units). Of the patients 51 (64%) experienced spontaneous resolution/improvement with a mean followup of 29 months before resolution, discharge home and/or end of followup. Only 20 patients (25%) underwent surgery. Cox regression analysis revealed that a normal initial dimercapto-succinic acid scan, initial reflux grade less than 5 and absent breakthrough febrile urinary tract infections were predictive of reflux resolution/improvement (p <0.05). Dimercapto-succinic acid scan abnormalities, prenatal hydronephrosis and breakthrough febrile urinary tract infections were significant predictors of surgery (p <0.05). Conclusions: Dilating vesicoureteral reflux in infancy often resolves/improves spontaneously. Therefore, surgery should be directed toward patients unlikely to experience resolution, ie those with an abnormal initial dimercapto-succinic acid scan, grade 5 vesicoureteral reflux and breakthrough febrile urinary tract infections. Key Words: urinary bladder, ureter, vesico-ureteral reflux, urinary tract infections, infant
HIGH grade VUR is considered a major risk factor for APN and renal scarring, which may precipitate hypertension and chronic renal failure. Dilating VUR (grades 3 to 5) was found in 9% to 30% of asymptomatic
1620
j
www.jurology.com
infants evaluated for prenatally detected hydronephrosis.1e3 Alternatively, dilating VUR was found in 12% to 16% of children with UTIs.4,5 While much emphasis is placed on identifying and preventing
0022-5347/14/1915-1620/0 THE JOURNAL OF UROLOGY® © 2014 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC.
http://dx.doi.org/10.1016/j.juro.2013.08.078 Vol. 191, 1620-1627, May 2014 Printed in U.S.A.
DILATING VESICOURETERAL REFLUX CAN BE TREATED NONOPERATIVELY
pyelonephritic renal damage, congenital causes of hypodysplasia should be considered. In 23% to 62% of cases congenital abnormalities are identified on DMSA scan, such as a small kidney and diffusely decreased uptake in the absence of a prior UTI,6,7 while pyelonephritic renal damage is segmental in most cases.8e10 Although not predominant, focal renal cortical abnormalities can also be seen without exposure to infection. The optimal balance between decreasing APN morbidity, including progressive renal injury and the avoidance of unnecessary treatment in children with VUR, is widely debated. Surgical treatment of severe reflux was traditionally advocated based on the assumption that resolution is unlikely and concern for APN mediated renal scarring. While the prior literature shows higher than expected rates of spontaneous resolution in infants with dilating VUR, there is a lack of randomized trials in infants showing whether medical or surgical management is best practice.11,12 Prognostic factors that predict long-term outcome are lacking. If these factors were better defined, a conservative management strategy for dilating reflux in infants may prevent unnecessary surgery. We report our results in infants with only dilating VUR that was managed by a nonoperative algorithm.
PATIENTS AND METHODS In a retrospective review using ICD-9 codes for VUR (593.70-73) we identified infants newly found to have VUR at our facility from January 2004 to January 2010, ensuring at least 2 years of followup. Patient charts were
1621
reviewed for age, sex, all renal imaging, urine cultures and all clinic/telephone encounters from the time of the initial visit until the end of the study period. Data were collected until patient discharge from clinic, surgery or end of the study period (fig. 1). Infants 6 months old or younger at study entry who had primary VUR grade 3 or greater (dilating VUR) on contrast VCUG according to the International Classification of Reflux were included in analysis and grouped by presentation.13 In 1 group infants presented with prenatal hydronephrosis and underwent DMSA renal scan before a UTI. In the other group infants presented after a single, culture positive febrile UTI, as documented by the pediatrician or laboratory results. All underwent DMSA scan before any subsequent UTI. All patients were on continuous prophylaxis with an antibiotic chosen by the treating physician. When laboratory results were available, bUTIs were defined as catheterized urine samples with greater than 50,000 cfu/ml in association with fever and pyuria while the child was on antibiotic prophylaxis, as documented by primary providers. When laboratory results were not available, the clinic notes of the pediatrician were relied on for the diagnosis. Surgery was discussed with the family when the child had 1 or more bUTIs, worsening VUR after 18-month followup VCUG, new renal scars or decreasing relative renal function on DMSA scans. Patients with VUR associated with a secondary cause, such as neuropathic bladder, posterior urethral valves or ureterocele, and those with nondilating VUR (grade 1 or 2) were excluded from study. Patients without an initial DMSA scan and at least 1 followup cystogram were also excluded. Patients with unilateral dilating VUR and contralateral nondilating VUR were included on patient level analysis but nondilating renal units were excluded from renal unit level analysis.
Figure 1. Study cohort breakdown
1622
DILATING VESICOURETERAL REFLUX CAN BE TREATED NONOPERATIVELY
A urologist reviewed all imaging to resolve any grade discordance between the treating urologist and radiologist reports. VUR was defined as resolved if it was not seen on contrast or radionuclide cystogram at last followup. Improvement was defined as a reduction to VUR grade less than 3. Worsening was defined as an increase of 1 or more grades from that on initial VCUG. Bilateral reflux was considered resolved if each ureter showed resolution. Followup radionuclide cystogram findings were converted to VCUG grades, including grade 1dreflux in the ureter only, grade 2dreflux into the pelvicalyceal system with no or minimal dilatation, grade 3dmoderate pelvicalyceal dilatation, grade 4dsevere pelvicalyceal dilatation without ureteral dilatation and grade 5dsevere dilatation of the pelvicalyceal system and ureter. Similar conversions were published previously.14,15 DMSA renal scan abnormalities were defined using a slight modification of RIVUR (Randomized Intervention for Children with Vesicoureteral Reflux) study criteria.16 The final classification resulted in 3 DMSA designations, including 1) focal renal scarring characterized by photon deficient defects with loss or contraction of the renal cortex, 2) acute pyelonephritis characterized by photon deficient defects with preservation of the renal contour and 3) a global hypotrophic/hypoplastic kidney without focal defects. Any of these 3 findings resulted in a DMSA classification of abnormal. Descriptive statistics were used to compare the 2 presentation groups. We used the Fisher exact test to compare the crude incidence of bUTIs among different groups. A Cox proportional hazards model was created to assess the association between early DMSA status, patient presentation, sex, circumcision status, VUR grade, bilateral reflux and bUTIs with VUR resolution and/or improvement, and the decision to proceed to surgery. All independent variables were treated as categorical. Analysis was performed to consider the clustered nature of the data, ie renal units in patients. Using significant factors we then created nomograms to predict the probability of resolution and/or improvement at 2 and 4 years. The nomograms were internally validated using 200 bootstrap samples with accuracy assessed using the Harrell c-statistic adapted for censored data.17 All analysis was done in SASÒ for WindowsÒ 9.2.
RESULTS Of the 90 infants eligible for study 80 (113 renal units) completed followup, including 34 (43%) who presented with prenatal hydronephrosis and 46 (57%) who presented after a febrile UTI (table 1). Patients presenting with prenatal hydronephrosis tended to be male and circumcised, and have a higher VUR grade (p <0.05). Age, overall followup, surgical rate, followup time to surgery, VUR laterality distribution and DMSA findings were similar in the 2 groups. Of infants with prenatal hydronephrosis 35% had DMSA scans with small kidneys and diffusely reduced uptake suggestive of congenital reflux nephropathy vs only 11% of those who
Table 1. Study cohort characteristics
No. pts No. male/female* % Circumcised males* Median mos followup (range) No. VUR correction surgery (%) Median mos to surgery (range) No. unilat/bilat VUR No. max VUR grade (%):* 3 4 5 No. DMSA results (%): Normal Focal scars Acute pyelonephritis Congenital reflux nephropathy Total abnormal
Prenatal Hydronephrosis
Febrile UTI
34 26/8 73 22 (1e76) 10 (29) 33 (1e76) 14/20
46 18/28 33 22 (9e93) 10 (22) 24 (9e49) 13/33
5 12 17
(15) (35) (50)
15 20 11
(33) (43) (24)
14 6 2 12
(41)
25 7† 10 5†
(54)
(59)
21
(46)
20
p Value e 0.001 0.014 0.197 0.447 0.734 0.243 0.036
0.267 0.770 0.062 0.012
* For group comparisons Wilcoxon rank sum test for continuous variables, and Fisher exact or chi-square test for discrete variables p <0.05. † Single patient with scarring and reflux nephropathy included in each group.
presented with a febrile UTI (p ¼ 0.01). The 10 patients not included in analysis were lost to followup before a second VCUG except one with grade 3 reflux that uneventfully improved to grade 1 within 2 years before clinic discharge. That patient did not undergo DMSA scan upon study enrollment. These 10 patients had an initial distribution of characteristics similar to that of the study cohort (data not shown). All patients were prescribed prophylactic antibiotics but compliance was not assessed. Most patients older than 2 months were prescribed sulfamethoxazole-trimethoprim (69%) vs nitrofurantoin (30%). Of the males 57% were circumcised and 70% of the males who experienced bUTIs were uncircumcised. The overall incidence of bUTIs was 35% with a crude rate of 0.23 UTIs per patientyear of followup. Only female sex was significantly associated with the incidence of any bUTI (p ¼ 0.02). Multivariable logistic regression of dilated renal units, including DMSA status, presentation, reflux grade, sex and bilateral reflux, also confirmed that only female sex was a risk factor for bUTIs (OR 4.4, 95% CI 1.8e11.0, p ¼ 0.001). Cox proportional hazards regression was used to determine predictors of VUR resolution and/or improvement. Only DMSA status, grade 5 VUR and a history of any bUTI were statistically significant in the model whether analyzed by patient or by renal unit (table 2). Figure 2 shows the probability of VUR resolution and/or improvement in these patients with time relative to initial VUR grade status. Overall 51 patients (64%) had spontaneous resolution/improvement with a mean followup of 29 months before resolution, discharge from care
DILATING VESICOURETERAL REFLUX CAN BE TREATED NONOPERATIVELY
1623
Table 2. Cox proportional hazards models of patient data with resolution or improvement and surgery as dependent variables Resolution/Improvement HR (95% CI) DMSA abnormality Any bUTI Initial VUR grade (5 vs referent 3e4) Presentation (referent febrile UTI) Uncircumcised (referent female) Circumcised (referent female)
0.50 0.31 0.27 0.74 1.17 1.25
(0.28e0.87) (0.17e0.58) (0.11e0.66) (0.42e1.29) (0.56e2.45) (0.71e2.20)
and/or end of followup. Resolution/improvement was seen in 90% of patients with normal vs 39% with abnormal DMSA scans at presentation (51% vs 15% complete resolution). Similarly, resolution/ improvement was noted in 77% of all patients who remained bUTI free vs 39% of those with at least 1 bUTI (46% vs 7% complete resolution). Nine patients had grade 3 or greater VUR at the conclusion of the study period at an average of 38 months of followup. All 9 patients started with grade 4 or 5 VUR and 8 improved at least 1 grade during followup. Four of these patients were discharged home with grade 3 VUR after at least 1 year free of UTI while off antibiotic prophylaxis and after having been toilet trained. Two patients failed to present for a recommended repeat VCUG, representing the third and fourth study in 1 each. Telephone calls to these patients 3 years later revealed no evidence of UTI while off prophylaxis. Two patients with improvement from grade 5 to 3 VUR relocated before the third study and could not be contacted. The last child, in whom initial grade 5 VUR improved to grade 3 after 18 months without further improvement, is still under our care after 4 years and is being treated for voiding dysfunction without UTIs on prophylaxis (fig. 3). The probability of complete resolution and/or improvement by initial VUR grade, DMSA status and history of bUTIs can be determined using the developed nomograms (fig. 4).
Surgery p Value 0.015 <0.001 0.004 0.284 0.672 0.435
HR (95% CI) 4.13 3.79 1.52 0.24 0.37 0.34
(1.78e9.51) (1.63e11.34) (0.75e2.85) (0.08e0.58) (0.14e1.04) (0.10e1.15)
p Value 0.001 0.003 0.263 0.002 0.059 0.082
Only 20 of the 80 patients (25%) underwent surgical correction. Surgery was indicated for 1 or more of certain reasons, including bUTIs in 65% of cases, persistent/worsening VUR after at least 12 months of observation in 65%, and/or decreased relative renal function or new scarring on DMSA scan in 60%. Interestingly, regression modeling for predictors of surgery revealed that abnormal DMSA status, prenatal hydronephrosis and any bUTI were each a statistically significant risk factor (p <0.05), although initial reflux grade was not.
DISCUSSION Eliminating antibiotic prophylaxis and surgery in children with low grade VUR (grade 2 or less) has cautiously gained more acceptance.18 For higher VUR grades concern for progressive renal damage drives some practitioners to recommend early surgery despite past evidence to the contrary.19 More recently, Szlyk et al noted only a 17% chance of new acute inflammatory defects on DMSA scan after a bUTI with 50% of patients having dilating VUR.20 Despite these data dilating VUR in infants is still occasionally treated with surgery.2,21 Assael et al examined 108 infants with prenatally detected VUR.2 In those with VUR grade 4 or greater 16% of renal units resolved spontaneously, 6% improved and 16% remained high grade. Of the patients 62% underwent reimplantation by a mean age of 30 months. In contrast, only 27% of the infants in
Figure 2. Cumulative incidence of resolution/improvement and complete resolution (blue bars) by initial VUR grade
1624
DILATING VESICOURETERAL REFLUX CAN BE TREATED NONOPERATIVELY
Figure 3. VUR outcome by initial VUR grade 1 (orange bars), 2 (green bars), 3 (yellow bars) and 4 (purple bars). Red bars indicate surgery.
our study with VUR grade 4 or greater underwent surgery with resolution or improvement in 58%. The discrepancy in the surgical rate among similar patient cohorts implies divergent thresholds for recommending repair, although this could not be confirmed due to the lack of surgical criteria reporting by Assael et al. As seen in this study and others, many more factors predict progression to surgery than have been found to influence resolution or improvement.22,23 Compared to infants diagnosed after UTI the prenatal group predominantly comprised males and/ or patients with a much higher reflux grade.6,24,25 Some investigators suggested that even in patients in whom reflux is diagnosed after a UTI a higher grade of VUR and the male sex predominate in the first year of life.26,27 In our cohort we confirmed a male sex predominance in the prenatal hydronephrosis group and a female sex predominance in the UTI group even in the first year of life. Infants with prenatal hydronephrosis also had higher grades of dilating reflux compared with the UTI group, as in past studies.6,24,25 While there was no association of patient presentation with VUR and overall resolution/improvement, presentation was associated with the decision to proceed to surgery in 29% of those with prenatal hydronephrosis vs 22% with febrile UTIs (Cox model p <0.05), which may reflect surgeon bias.
Literature supports the notion that initial DMSA scan status predicts VUR resolution. In a study of 236 refluxing renal units the overall VUR resolution rate in 193 that were evaluable at 15 months of followup was 70% of those with an initially normal DMSA scan compared to only 33% with an initially abnormal DMSA scan.28 The resolution rate of grades 4 to 5 VUR was 43% (35 of 81 renal units). In another series Godley et al examined 40 infants 7 months old or younger with grade 3 or greater primary VUR on VCUG within a 2-year period.29 Of the patients 29 had prenatal hydronephrosis and 10 presented with a UTI. The group analyzed VUR resolution and bladder function at a mean patient age of 16 months. Reflux resolved in all 14 patients with normal kidneys compared to only 8 of 26 (31%, 95% CI 14e52) with abnormal kidneys bilaterally or unilaterally. Others reported similar findings in patients younger than 1 year.12,30 Our findings further substantiate these observations. As expected, bUTI was not only inversely associated with the resolution rate but also positively associated with progression to surgery. This was expected since bUTI was used as the clinical indicator to reevaluate the patient treatment pathway. As mentioned, we did not assess antibiotic compliance but our UTI rate was quite low and similar to that in other comparable published cohorts.2 Notably, only 50% of bUTIs reported by pediatricians could be confirmed by an available laboratory report. Due to the lack of compliance data and the low overall infection rate it is difficult to confidently assert which patients were at higher risk for bUTI in our series. This study is limited by its retrospective nature and the cohort size, especially when considering subgroup analysis. However, it still represents one of the largest cohorts of infants with only dilating VUR assessed to date. Its real value is the demonstration that an initial nonoperative algorithm can result in reasonable natural resolution or improvement of dilating VUR in infants with a relatively low rate of surgical intervention. In acknowledgment of possible criticisms we mention that we discharge some children with unresolved VUR when significant improvement has been seen (VUR grade 3 or less) and there has been a proven period (greater than 6 months) without UTI while off antibiotic prophylaxis after toilet training is completed. Based on prior literature we argue that in the absence of UTI new, significant renal injury would be highly unlikely, although we do not have followup DMSA scans to confirm this in our cohort. Similarly, we cannot confirm in all patients that they did not present to the primary physician later with recurrent UTIs after discharge from our care.
DILATING VESICOURETERAL REFLUX CAN BE TREATED NONOPERATIVELY
1625
Figure 4. Nomograms to predict probability at 2 and 4 years of followup. For each clinical characteristic find number of corresponding points by drawing line from characteristic to uppermost “Points” scale. Sum points for each characteristic. Find corresponding number on “Total points” scale and draw line down to corresponding 2 and 4-year probabilities. For example, infant with grade 5 VUR þ bUTI þ abnormal DMSA ¼ 100 þ 100 þ 50 ¼ 250 total points ¼ 2% 2-year probability of complete resolution. A, complete resolution. B, complete resolution and/or improvement.
One could argue that direct evidence showing that any VUR treatment protects the kidneys in the long term is limited at best. However, we believe that the amount of indirect evidence signals caution to any who propose complete ignorance. Finally, we have no data on the presence or absence of bladder-bowel dysfunction and, therefore, we
cannot comment on its effect on spontaneous resolution in our cohort.
CONCLUSIONS These data lend further support to the high rate of spontaneous improvement/resolution in infants
1626
DILATING VESICOURETERAL REFLUX CAN BE TREATED NONOPERATIVELY
with dilating VUR, the importance of initial DMSA scan status in management decisions and the validity of nonsurgical treatment when bUTIs are controlled with continuous antibiotic prophylaxis.
These patients certainly still require close followup and good urologist-to-pediatrician communication to be certain that they remain healthy going forward.
REFERENCES 1. Blachar A, Blachar Y, Livne PM et al: Clinical outcome and follow-up of prenatal hydronephrosis. Pediatr Nephrol 1994; 8: 30.
prenatal hydronephrosis: long-term results of a prospective study. J Urol 2003; 169: 1837.
2. Assael BM, Guez S, Marra G et al: Congenital reflux nephropathy: a follow-up of 108 cases diagnosed perinatally. Br J Urol 1998; 82: 252.
12. Sj€ostr€om S, Sillen U, Jodal U et al: Predictive factors for resolution of congenital high grade vesicoureteral reflux in infants: results of univariate and multivariate analyses. J Urol 2010; 183: 1177.
3. Tam JC, Hodson EM, Choong KK et al: Postnatal diagnosis and outcome of urinary tract abnormalities detected by antenatal ultrasound. Med J Aust 1994; 160: 633.
13. Lebowitz RL, Olbing H, Parkkulainen KV et al: International system of radiographic grading of vesicoureteric reflux. International Reflux Study in Children. Pediatr Radiol 1985; 15: 105.
4. Hansson S, Dhamey M, Sigstrom O et al: Dimercapto-succinic acid scintigraphy instead of voiding cystourethrography for infants with urinary tract infection. J Urol 2004; 172: 1071.
14. Polito C, Rambaldi PF, La Manna A et al: Enhanced detection of vesicoureteric reflux with isotopic cystography. Pediatr Nephrol 2000; 14: 827.
5. Hansson S, Bollgren I, Esbj€orner E et al: Urinary tract infections in children below two years of age: a quality assurance project in Sweden. The Swedish Pediatric Nephrology Association. Acta Paediatr 1999; 88: 270. 6. Anderson PA and Rickwood AM: Features of primary vesicoureteric reflux detected by prenatal sonography. Br J Urol 1991; 67: 267. 7. Oliveira EA, Diniz JS, Silva JM et al: Features of primary vesicoureteric reflux detected by investigation of foetal hydronephrosis. Int Urol Nephrol 1998; 30: 535.
15. Unver T, Alpay H, Biyikli NK et al: Comparison of direct radionuclide cystography and voiding cystourethrography in detecting vesicoureteral reflux. Pediatr Int 2006; 48: 287.
22. Ogan K, Pohl HG, Carlson D et al: Parental preferences in the management of vesicoureteral reflux. J Urol 2001; 166: 240. 23. Szymanski KM, Oliveira LM, Silva A et al: Analysis of indications for ureteral reimplantation in 3738 children with vesicoureteral reflux: a single institutional cohort. J Pediatr Urol 2011; 7: 601. 24. Lenaghan D, Whitaker JG, Jensen F et al: The natural history of reflux and long-term effects of reflux on the kidney. J Urol 1976; 115: 728.
16. Ziessman HA and Majd M: Importance of methodology on (99m)technetium dimercapto-succinic acid scintigraphic image quality: imaging pilot study for RIVUR (Randomized Intervention for Children With Vesicoureteral Reflux) multicenter investigation. J Urol 2009; 182: 272.
25. Najmaldin A, Burge DM and Atwell JD: Fetal vesicoureteric reflux. Br J Urol 1990; 65: 403.
17. Pencina MJ, D’Agostino RB Sr and Song L: Quantifying discrimination of Framingham risk functions with different survival C statistics. Stat Med 2012; 31: 1543.
27. Goldraich NP and Goldraich IH: Followup of conservatively treated children with high and low grade vesicoureteral reflux: a prospective study. J Urol 1992; 148: 1688.
8. Rushton HG, Majd M, Jantausch B et al: Renal scarring following reflux and nonreflux pyelonephritis in children: evaluation with 99mtechnetium-dimercaptosuccinic acid scintigraphy. J Urol 1992; 147: 1327.
18. Peters CA, Skoog SJ, Arant BS et al: Summary of the AUA Guideline on Management of Primary Vesicoureteral Reflux in Children. J Urol 2010; 184: 1134.
9. Sty JR, Wells RG, Starshak RJ et al: Imaging in acute renal infection in children. AJR Am J Roentgenol 1987; 148: 471.
19. Skoog SJ, Belman AB and Majd M: A nonsurgical approach to the management of primary vesicoureteral reflux. J Urol 1987; 138: 941.
10. Traisman ES, Conway JJ, Traisman HS et al: The localization of urinary tract infection with 99mTc glucoheptonate scintigraphy. Pediatr Radiol 1986; 16: 403.
20. Szlyk GR, Williams SB, Majd M et al: Incidence of new renal parenchymal inflammatory changes following breakthrough urinary tract infection in patients with vesicoureteral reflux treated with antibiotic prophylaxis: evaluation by 99MTechnetium dimercapto-succinic acid renal scan. J Urol 2003; 170: 1566.
11. Upadhyay J, McLorie GA, Bolduc S et al: Natural history of neonatal reflux associated with
21. Puri P, Kutasy B, Colhoun E et al: Single center experience with endoscopic subureteral dextranomer/hyaluronic acid injection as first line treatment in 1,551 children with intermediate and high grade vesicoureteral reflux. J Urol 2012; 188: 1485.
26. Rolleston GL, Shannon FT and Utley WL: Followup of vesico-ureteric reflux in the newborn. Kidney Int Suppl 1975; 4: S59.
28. Yeung CK, Godley ML, Dhillon HK et al: The characteristics of primary vesico-ureteric reflux in male and female infants with pre-natal hydronephrosis. Br J Urol 1997; 80: 319. 29. Godley ML, Desai D, Yeung CK et al: The relationship between early renal status, and the resolution of vesico-ureteric reflux and bladder function at 16 months. BJU Int 2001; 87: 457. 30. Nakamura M, Moriya K, Mitsui T et al: Abnormal dimercapto-succinic acid scan is a predictive factor of breakthrough urinary tract infection in children with primary vesicoureteral reflux. J Urol 2009; 182: 1694.
EDITORIAL COMMENT The authors present a relatively large series of infants with high grade reflux demonstrating that most could be treated nonoperatively. Indications for surgery were bUTI, worsening VUR grade on 18-month followup VCUG and the appearance of
new renal scars or decreasing relative renal function on DMSA scans. They report efforts to decrease the rate of surgical intervention in this patient group. However, one wonders whether the envelope could be pushed even
DILATING VESICOURETERAL REFLUX CAN BE TREATED NONOPERATIVELY
further. In the absence of infections do these patients require followup VCUG at 18 months? From a renal standpoint is there benefit in correcting persistent VUR? Is bUTI the real game changer? If so, our efforts should be focused on preventing infection by correcting bladder and bowel dysfunction. This would not only decrease the surgical intervention rate from the current 25% but also reduce the morbidity and cost of followup without placing the kidneys at risk. This approach obviously mandates responsible families and effective communication between the urologist and pediatrician. The same argument applies to noninfection related decreasing relative renal function on DMSA scans. The increase in differential function of the unaffected kidney may simply reflect compensatory growth. In the absence of infection
1627
there is no evidence that correcting VUR results in catch-up growth of the affected kidney and reversal of the negative trend in relative renal function. Many questions in VUR can only be answered by objective data from prospective multicenter trials. Only then will we achieve our goal of minimizing the morbidity of treatment and followup while preventing renal injury. Antoine Khoury Department of Urology University of California-Irvine Irvine, California and Pediatric Urology Children’s Hospital of Orange County Orange, California
Abbreviations and Acronyms APN ¼ acute pyelonephritis bUTI ¼ breakthrough febrile UTI DMSA ¼ dimercapto-succinic acid UTI ¼ urinary tract infection VCUG ¼ voiding cystourethrogram VUR ¼ vesicoureteral reflux Accepted for publication August 26, 2013. Study received Children’s National Medical Center institutional review board approval. * Correspondence: Louisiana State University Health Sciences Center and Children’s Hospital New Orleans, Department of Urology, 200 Henry Clay Ave., New Orleans, Louisiana 70118 (telephone: 504-896-9233; FAX: 504-896-9861; e-mail: [email protected]).
Purpose: Dilating vesicoureteral reflux provokes concern for physicians and parents that often leads to corrective surgery in young children. Since there are limited data describing the natural history of dilating vesicoureteral reflux in infants, we identified factors predictive of resolution/improvement in infants initially treated nonoperatively. Materials and Methods: We reviewed the medical records of 90 infants 6 months old or younger from 2004 to 2010 who were referred for prenatal hydronephrosis or initial febrile urinary tract infection and found to have dilating vesicoureteral reflux (grade 3 or greater). Variables of interest included presentation, dimercapto-succinic acid results, sex, breakthrough febrile urinary tract infections, reflux grade and bilateral reflux. Cox regression analysis was performed to determine predictors of spontaneous resolution and/or improvement to reflux grade less than 3 as well as predictors of surgical intervention. Results: Included in final analysis were 80 infants (113 renal units). Of the patients 51 (64%) experienced spontaneous resolution/improvement with a mean followup of 29 months before resolution, discharge home and/or end of followup. Only 20 patients (25%) underwent surgery. Cox regression analysis revealed that a normal initial dimercapto-succinic acid scan, initial reflux grade less than 5 and absent breakthrough febrile urinary tract infections were predictive of reflux resolution/improvement (p <0.05). Dimercapto-succinic acid scan abnormalities, prenatal hydronephrosis and breakthrough febrile urinary tract infections were significant predictors of surgery (p <0.05). Conclusions: Dilating vesicoureteral reflux in infancy often resolves/improves spontaneously. Therefore, surgery should be directed toward patients unlikely to experience resolution, ie those with an abnormal initial dimercapto-succinic acid scan, grade 5 vesicoureteral reflux and breakthrough febrile urinary tract infections. Key Words: urinary bladder, ureter, vesico-ureteral reflux, urinary tract infections, infant
HIGH grade VUR is considered a major risk factor for APN and renal scarring, which may precipitate hypertension and chronic renal failure. Dilating VUR (grades 3 to 5) was found in 9% to 30% of asymptomatic
1620
j
www.jurology.com
infants evaluated for prenatally detected hydronephrosis.1e3 Alternatively, dilating VUR was found in 12% to 16% of children with UTIs.4,5 While much emphasis is placed on identifying and preventing
0022-5347/14/1915-1620/0 THE JOURNAL OF UROLOGY® © 2014 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC.
http://dx.doi.org/10.1016/j.juro.2013.08.078 Vol. 191, 1620-1627, May 2014 Printed in U.S.A.
DILATING VESICOURETERAL REFLUX CAN BE TREATED NONOPERATIVELY
pyelonephritic renal damage, congenital causes of hypodysplasia should be considered. In 23% to 62% of cases congenital abnormalities are identified on DMSA scan, such as a small kidney and diffusely decreased uptake in the absence of a prior UTI,6,7 while pyelonephritic renal damage is segmental in most cases.8e10 Although not predominant, focal renal cortical abnormalities can also be seen without exposure to infection. The optimal balance between decreasing APN morbidity, including progressive renal injury and the avoidance of unnecessary treatment in children with VUR, is widely debated. Surgical treatment of severe reflux was traditionally advocated based on the assumption that resolution is unlikely and concern for APN mediated renal scarring. While the prior literature shows higher than expected rates of spontaneous resolution in infants with dilating VUR, there is a lack of randomized trials in infants showing whether medical or surgical management is best practice.11,12 Prognostic factors that predict long-term outcome are lacking. If these factors were better defined, a conservative management strategy for dilating reflux in infants may prevent unnecessary surgery. We report our results in infants with only dilating VUR that was managed by a nonoperative algorithm.
PATIENTS AND METHODS In a retrospective review using ICD-9 codes for VUR (593.70-73) we identified infants newly found to have VUR at our facility from January 2004 to January 2010, ensuring at least 2 years of followup. Patient charts were
1621
reviewed for age, sex, all renal imaging, urine cultures and all clinic/telephone encounters from the time of the initial visit until the end of the study period. Data were collected until patient discharge from clinic, surgery or end of the study period (fig. 1). Infants 6 months old or younger at study entry who had primary VUR grade 3 or greater (dilating VUR) on contrast VCUG according to the International Classification of Reflux were included in analysis and grouped by presentation.13 In 1 group infants presented with prenatal hydronephrosis and underwent DMSA renal scan before a UTI. In the other group infants presented after a single, culture positive febrile UTI, as documented by the pediatrician or laboratory results. All underwent DMSA scan before any subsequent UTI. All patients were on continuous prophylaxis with an antibiotic chosen by the treating physician. When laboratory results were available, bUTIs were defined as catheterized urine samples with greater than 50,000 cfu/ml in association with fever and pyuria while the child was on antibiotic prophylaxis, as documented by primary providers. When laboratory results were not available, the clinic notes of the pediatrician were relied on for the diagnosis. Surgery was discussed with the family when the child had 1 or more bUTIs, worsening VUR after 18-month followup VCUG, new renal scars or decreasing relative renal function on DMSA scans. Patients with VUR associated with a secondary cause, such as neuropathic bladder, posterior urethral valves or ureterocele, and those with nondilating VUR (grade 1 or 2) were excluded from study. Patients without an initial DMSA scan and at least 1 followup cystogram were also excluded. Patients with unilateral dilating VUR and contralateral nondilating VUR were included on patient level analysis but nondilating renal units were excluded from renal unit level analysis.
Figure 1. Study cohort breakdown
1622
DILATING VESICOURETERAL REFLUX CAN BE TREATED NONOPERATIVELY
A urologist reviewed all imaging to resolve any grade discordance between the treating urologist and radiologist reports. VUR was defined as resolved if it was not seen on contrast or radionuclide cystogram at last followup. Improvement was defined as a reduction to VUR grade less than 3. Worsening was defined as an increase of 1 or more grades from that on initial VCUG. Bilateral reflux was considered resolved if each ureter showed resolution. Followup radionuclide cystogram findings were converted to VCUG grades, including grade 1dreflux in the ureter only, grade 2dreflux into the pelvicalyceal system with no or minimal dilatation, grade 3dmoderate pelvicalyceal dilatation, grade 4dsevere pelvicalyceal dilatation without ureteral dilatation and grade 5dsevere dilatation of the pelvicalyceal system and ureter. Similar conversions were published previously.14,15 DMSA renal scan abnormalities were defined using a slight modification of RIVUR (Randomized Intervention for Children with Vesicoureteral Reflux) study criteria.16 The final classification resulted in 3 DMSA designations, including 1) focal renal scarring characterized by photon deficient defects with loss or contraction of the renal cortex, 2) acute pyelonephritis characterized by photon deficient defects with preservation of the renal contour and 3) a global hypotrophic/hypoplastic kidney without focal defects. Any of these 3 findings resulted in a DMSA classification of abnormal. Descriptive statistics were used to compare the 2 presentation groups. We used the Fisher exact test to compare the crude incidence of bUTIs among different groups. A Cox proportional hazards model was created to assess the association between early DMSA status, patient presentation, sex, circumcision status, VUR grade, bilateral reflux and bUTIs with VUR resolution and/or improvement, and the decision to proceed to surgery. All independent variables were treated as categorical. Analysis was performed to consider the clustered nature of the data, ie renal units in patients. Using significant factors we then created nomograms to predict the probability of resolution and/or improvement at 2 and 4 years. The nomograms were internally validated using 200 bootstrap samples with accuracy assessed using the Harrell c-statistic adapted for censored data.17 All analysis was done in SASÒ for WindowsÒ 9.2.
RESULTS Of the 90 infants eligible for study 80 (113 renal units) completed followup, including 34 (43%) who presented with prenatal hydronephrosis and 46 (57%) who presented after a febrile UTI (table 1). Patients presenting with prenatal hydronephrosis tended to be male and circumcised, and have a higher VUR grade (p <0.05). Age, overall followup, surgical rate, followup time to surgery, VUR laterality distribution and DMSA findings were similar in the 2 groups. Of infants with prenatal hydronephrosis 35% had DMSA scans with small kidneys and diffusely reduced uptake suggestive of congenital reflux nephropathy vs only 11% of those who
Table 1. Study cohort characteristics
No. pts No. male/female* % Circumcised males* Median mos followup (range) No. VUR correction surgery (%) Median mos to surgery (range) No. unilat/bilat VUR No. max VUR grade (%):* 3 4 5 No. DMSA results (%): Normal Focal scars Acute pyelonephritis Congenital reflux nephropathy Total abnormal
Prenatal Hydronephrosis
Febrile UTI
34 26/8 73 22 (1e76) 10 (29) 33 (1e76) 14/20
46 18/28 33 22 (9e93) 10 (22) 24 (9e49) 13/33
5 12 17
(15) (35) (50)
15 20 11
(33) (43) (24)
14 6 2 12
(41)
25 7† 10 5†
(54)
(59)
21
(46)
20
p Value e 0.001 0.014 0.197 0.447 0.734 0.243 0.036
0.267 0.770 0.062 0.012
* For group comparisons Wilcoxon rank sum test for continuous variables, and Fisher exact or chi-square test for discrete variables p <0.05. † Single patient with scarring and reflux nephropathy included in each group.
presented with a febrile UTI (p ¼ 0.01). The 10 patients not included in analysis were lost to followup before a second VCUG except one with grade 3 reflux that uneventfully improved to grade 1 within 2 years before clinic discharge. That patient did not undergo DMSA scan upon study enrollment. These 10 patients had an initial distribution of characteristics similar to that of the study cohort (data not shown). All patients were prescribed prophylactic antibiotics but compliance was not assessed. Most patients older than 2 months were prescribed sulfamethoxazole-trimethoprim (69%) vs nitrofurantoin (30%). Of the males 57% were circumcised and 70% of the males who experienced bUTIs were uncircumcised. The overall incidence of bUTIs was 35% with a crude rate of 0.23 UTIs per patientyear of followup. Only female sex was significantly associated with the incidence of any bUTI (p ¼ 0.02). Multivariable logistic regression of dilated renal units, including DMSA status, presentation, reflux grade, sex and bilateral reflux, also confirmed that only female sex was a risk factor for bUTIs (OR 4.4, 95% CI 1.8e11.0, p ¼ 0.001). Cox proportional hazards regression was used to determine predictors of VUR resolution and/or improvement. Only DMSA status, grade 5 VUR and a history of any bUTI were statistically significant in the model whether analyzed by patient or by renal unit (table 2). Figure 2 shows the probability of VUR resolution and/or improvement in these patients with time relative to initial VUR grade status. Overall 51 patients (64%) had spontaneous resolution/improvement with a mean followup of 29 months before resolution, discharge from care
DILATING VESICOURETERAL REFLUX CAN BE TREATED NONOPERATIVELY
1623
Table 2. Cox proportional hazards models of patient data with resolution or improvement and surgery as dependent variables Resolution/Improvement HR (95% CI) DMSA abnormality Any bUTI Initial VUR grade (5 vs referent 3e4) Presentation (referent febrile UTI) Uncircumcised (referent female) Circumcised (referent female)
0.50 0.31 0.27 0.74 1.17 1.25
(0.28e0.87) (0.17e0.58) (0.11e0.66) (0.42e1.29) (0.56e2.45) (0.71e2.20)
and/or end of followup. Resolution/improvement was seen in 90% of patients with normal vs 39% with abnormal DMSA scans at presentation (51% vs 15% complete resolution). Similarly, resolution/ improvement was noted in 77% of all patients who remained bUTI free vs 39% of those with at least 1 bUTI (46% vs 7% complete resolution). Nine patients had grade 3 or greater VUR at the conclusion of the study period at an average of 38 months of followup. All 9 patients started with grade 4 or 5 VUR and 8 improved at least 1 grade during followup. Four of these patients were discharged home with grade 3 VUR after at least 1 year free of UTI while off antibiotic prophylaxis and after having been toilet trained. Two patients failed to present for a recommended repeat VCUG, representing the third and fourth study in 1 each. Telephone calls to these patients 3 years later revealed no evidence of UTI while off prophylaxis. Two patients with improvement from grade 5 to 3 VUR relocated before the third study and could not be contacted. The last child, in whom initial grade 5 VUR improved to grade 3 after 18 months without further improvement, is still under our care after 4 years and is being treated for voiding dysfunction without UTIs on prophylaxis (fig. 3). The probability of complete resolution and/or improvement by initial VUR grade, DMSA status and history of bUTIs can be determined using the developed nomograms (fig. 4).
Surgery p Value 0.015 <0.001 0.004 0.284 0.672 0.435
HR (95% CI) 4.13 3.79 1.52 0.24 0.37 0.34
(1.78e9.51) (1.63e11.34) (0.75e2.85) (0.08e0.58) (0.14e1.04) (0.10e1.15)
p Value 0.001 0.003 0.263 0.002 0.059 0.082
Only 20 of the 80 patients (25%) underwent surgical correction. Surgery was indicated for 1 or more of certain reasons, including bUTIs in 65% of cases, persistent/worsening VUR after at least 12 months of observation in 65%, and/or decreased relative renal function or new scarring on DMSA scan in 60%. Interestingly, regression modeling for predictors of surgery revealed that abnormal DMSA status, prenatal hydronephrosis and any bUTI were each a statistically significant risk factor (p <0.05), although initial reflux grade was not.
DISCUSSION Eliminating antibiotic prophylaxis and surgery in children with low grade VUR (grade 2 or less) has cautiously gained more acceptance.18 For higher VUR grades concern for progressive renal damage drives some practitioners to recommend early surgery despite past evidence to the contrary.19 More recently, Szlyk et al noted only a 17% chance of new acute inflammatory defects on DMSA scan after a bUTI with 50% of patients having dilating VUR.20 Despite these data dilating VUR in infants is still occasionally treated with surgery.2,21 Assael et al examined 108 infants with prenatally detected VUR.2 In those with VUR grade 4 or greater 16% of renal units resolved spontaneously, 6% improved and 16% remained high grade. Of the patients 62% underwent reimplantation by a mean age of 30 months. In contrast, only 27% of the infants in
Figure 2. Cumulative incidence of resolution/improvement and complete resolution (blue bars) by initial VUR grade
1624
DILATING VESICOURETERAL REFLUX CAN BE TREATED NONOPERATIVELY
Figure 3. VUR outcome by initial VUR grade 1 (orange bars), 2 (green bars), 3 (yellow bars) and 4 (purple bars). Red bars indicate surgery.
our study with VUR grade 4 or greater underwent surgery with resolution or improvement in 58%. The discrepancy in the surgical rate among similar patient cohorts implies divergent thresholds for recommending repair, although this could not be confirmed due to the lack of surgical criteria reporting by Assael et al. As seen in this study and others, many more factors predict progression to surgery than have been found to influence resolution or improvement.22,23 Compared to infants diagnosed after UTI the prenatal group predominantly comprised males and/ or patients with a much higher reflux grade.6,24,25 Some investigators suggested that even in patients in whom reflux is diagnosed after a UTI a higher grade of VUR and the male sex predominate in the first year of life.26,27 In our cohort we confirmed a male sex predominance in the prenatal hydronephrosis group and a female sex predominance in the UTI group even in the first year of life. Infants with prenatal hydronephrosis also had higher grades of dilating reflux compared with the UTI group, as in past studies.6,24,25 While there was no association of patient presentation with VUR and overall resolution/improvement, presentation was associated with the decision to proceed to surgery in 29% of those with prenatal hydronephrosis vs 22% with febrile UTIs (Cox model p <0.05), which may reflect surgeon bias.
Literature supports the notion that initial DMSA scan status predicts VUR resolution. In a study of 236 refluxing renal units the overall VUR resolution rate in 193 that were evaluable at 15 months of followup was 70% of those with an initially normal DMSA scan compared to only 33% with an initially abnormal DMSA scan.28 The resolution rate of grades 4 to 5 VUR was 43% (35 of 81 renal units). In another series Godley et al examined 40 infants 7 months old or younger with grade 3 or greater primary VUR on VCUG within a 2-year period.29 Of the patients 29 had prenatal hydronephrosis and 10 presented with a UTI. The group analyzed VUR resolution and bladder function at a mean patient age of 16 months. Reflux resolved in all 14 patients with normal kidneys compared to only 8 of 26 (31%, 95% CI 14e52) with abnormal kidneys bilaterally or unilaterally. Others reported similar findings in patients younger than 1 year.12,30 Our findings further substantiate these observations. As expected, bUTI was not only inversely associated with the resolution rate but also positively associated with progression to surgery. This was expected since bUTI was used as the clinical indicator to reevaluate the patient treatment pathway. As mentioned, we did not assess antibiotic compliance but our UTI rate was quite low and similar to that in other comparable published cohorts.2 Notably, only 50% of bUTIs reported by pediatricians could be confirmed by an available laboratory report. Due to the lack of compliance data and the low overall infection rate it is difficult to confidently assert which patients were at higher risk for bUTI in our series. This study is limited by its retrospective nature and the cohort size, especially when considering subgroup analysis. However, it still represents one of the largest cohorts of infants with only dilating VUR assessed to date. Its real value is the demonstration that an initial nonoperative algorithm can result in reasonable natural resolution or improvement of dilating VUR in infants with a relatively low rate of surgical intervention. In acknowledgment of possible criticisms we mention that we discharge some children with unresolved VUR when significant improvement has been seen (VUR grade 3 or less) and there has been a proven period (greater than 6 months) without UTI while off antibiotic prophylaxis after toilet training is completed. Based on prior literature we argue that in the absence of UTI new, significant renal injury would be highly unlikely, although we do not have followup DMSA scans to confirm this in our cohort. Similarly, we cannot confirm in all patients that they did not present to the primary physician later with recurrent UTIs after discharge from our care.
DILATING VESICOURETERAL REFLUX CAN BE TREATED NONOPERATIVELY
1625
Figure 4. Nomograms to predict probability at 2 and 4 years of followup. For each clinical characteristic find number of corresponding points by drawing line from characteristic to uppermost “Points” scale. Sum points for each characteristic. Find corresponding number on “Total points” scale and draw line down to corresponding 2 and 4-year probabilities. For example, infant with grade 5 VUR þ bUTI þ abnormal DMSA ¼ 100 þ 100 þ 50 ¼ 250 total points ¼ 2% 2-year probability of complete resolution. A, complete resolution. B, complete resolution and/or improvement.
One could argue that direct evidence showing that any VUR treatment protects the kidneys in the long term is limited at best. However, we believe that the amount of indirect evidence signals caution to any who propose complete ignorance. Finally, we have no data on the presence or absence of bladder-bowel dysfunction and, therefore, we
cannot comment on its effect on spontaneous resolution in our cohort.
CONCLUSIONS These data lend further support to the high rate of spontaneous improvement/resolution in infants
1626
DILATING VESICOURETERAL REFLUX CAN BE TREATED NONOPERATIVELY
with dilating VUR, the importance of initial DMSA scan status in management decisions and the validity of nonsurgical treatment when bUTIs are controlled with continuous antibiotic prophylaxis.
These patients certainly still require close followup and good urologist-to-pediatrician communication to be certain that they remain healthy going forward.
REFERENCES 1. Blachar A, Blachar Y, Livne PM et al: Clinical outcome and follow-up of prenatal hydronephrosis. Pediatr Nephrol 1994; 8: 30.
prenatal hydronephrosis: long-term results of a prospective study. J Urol 2003; 169: 1837.
2. Assael BM, Guez S, Marra G et al: Congenital reflux nephropathy: a follow-up of 108 cases diagnosed perinatally. Br J Urol 1998; 82: 252.
12. Sj€ostr€om S, Sillen U, Jodal U et al: Predictive factors for resolution of congenital high grade vesicoureteral reflux in infants: results of univariate and multivariate analyses. J Urol 2010; 183: 1177.
3. Tam JC, Hodson EM, Choong KK et al: Postnatal diagnosis and outcome of urinary tract abnormalities detected by antenatal ultrasound. Med J Aust 1994; 160: 633.
13. Lebowitz RL, Olbing H, Parkkulainen KV et al: International system of radiographic grading of vesicoureteric reflux. International Reflux Study in Children. Pediatr Radiol 1985; 15: 105.
4. Hansson S, Dhamey M, Sigstrom O et al: Dimercapto-succinic acid scintigraphy instead of voiding cystourethrography for infants with urinary tract infection. J Urol 2004; 172: 1071.
14. Polito C, Rambaldi PF, La Manna A et al: Enhanced detection of vesicoureteric reflux with isotopic cystography. Pediatr Nephrol 2000; 14: 827.
5. Hansson S, Bollgren I, Esbj€orner E et al: Urinary tract infections in children below two years of age: a quality assurance project in Sweden. The Swedish Pediatric Nephrology Association. Acta Paediatr 1999; 88: 270. 6. Anderson PA and Rickwood AM: Features of primary vesicoureteric reflux detected by prenatal sonography. Br J Urol 1991; 67: 267. 7. Oliveira EA, Diniz JS, Silva JM et al: Features of primary vesicoureteric reflux detected by investigation of foetal hydronephrosis. Int Urol Nephrol 1998; 30: 535.
15. Unver T, Alpay H, Biyikli NK et al: Comparison of direct radionuclide cystography and voiding cystourethrography in detecting vesicoureteral reflux. Pediatr Int 2006; 48: 287.
22. Ogan K, Pohl HG, Carlson D et al: Parental preferences in the management of vesicoureteral reflux. J Urol 2001; 166: 240. 23. Szymanski KM, Oliveira LM, Silva A et al: Analysis of indications for ureteral reimplantation in 3738 children with vesicoureteral reflux: a single institutional cohort. J Pediatr Urol 2011; 7: 601. 24. Lenaghan D, Whitaker JG, Jensen F et al: The natural history of reflux and long-term effects of reflux on the kidney. J Urol 1976; 115: 728.
16. Ziessman HA and Majd M: Importance of methodology on (99m)technetium dimercapto-succinic acid scintigraphic image quality: imaging pilot study for RIVUR (Randomized Intervention for Children With Vesicoureteral Reflux) multicenter investigation. J Urol 2009; 182: 272.
25. Najmaldin A, Burge DM and Atwell JD: Fetal vesicoureteric reflux. Br J Urol 1990; 65: 403.
17. Pencina MJ, D’Agostino RB Sr and Song L: Quantifying discrimination of Framingham risk functions with different survival C statistics. Stat Med 2012; 31: 1543.
27. Goldraich NP and Goldraich IH: Followup of conservatively treated children with high and low grade vesicoureteral reflux: a prospective study. J Urol 1992; 148: 1688.
8. Rushton HG, Majd M, Jantausch B et al: Renal scarring following reflux and nonreflux pyelonephritis in children: evaluation with 99mtechnetium-dimercaptosuccinic acid scintigraphy. J Urol 1992; 147: 1327.
18. Peters CA, Skoog SJ, Arant BS et al: Summary of the AUA Guideline on Management of Primary Vesicoureteral Reflux in Children. J Urol 2010; 184: 1134.
9. Sty JR, Wells RG, Starshak RJ et al: Imaging in acute renal infection in children. AJR Am J Roentgenol 1987; 148: 471.
19. Skoog SJ, Belman AB and Majd M: A nonsurgical approach to the management of primary vesicoureteral reflux. J Urol 1987; 138: 941.
10. Traisman ES, Conway JJ, Traisman HS et al: The localization of urinary tract infection with 99mTc glucoheptonate scintigraphy. Pediatr Radiol 1986; 16: 403.
20. Szlyk GR, Williams SB, Majd M et al: Incidence of new renal parenchymal inflammatory changes following breakthrough urinary tract infection in patients with vesicoureteral reflux treated with antibiotic prophylaxis: evaluation by 99MTechnetium dimercapto-succinic acid renal scan. J Urol 2003; 170: 1566.
11. Upadhyay J, McLorie GA, Bolduc S et al: Natural history of neonatal reflux associated with
21. Puri P, Kutasy B, Colhoun E et al: Single center experience with endoscopic subureteral dextranomer/hyaluronic acid injection as first line treatment in 1,551 children with intermediate and high grade vesicoureteral reflux. J Urol 2012; 188: 1485.
26. Rolleston GL, Shannon FT and Utley WL: Followup of vesico-ureteric reflux in the newborn. Kidney Int Suppl 1975; 4: S59.
28. Yeung CK, Godley ML, Dhillon HK et al: The characteristics of primary vesico-ureteric reflux in male and female infants with pre-natal hydronephrosis. Br J Urol 1997; 80: 319. 29. Godley ML, Desai D, Yeung CK et al: The relationship between early renal status, and the resolution of vesico-ureteric reflux and bladder function at 16 months. BJU Int 2001; 87: 457. 30. Nakamura M, Moriya K, Mitsui T et al: Abnormal dimercapto-succinic acid scan is a predictive factor of breakthrough urinary tract infection in children with primary vesicoureteral reflux. J Urol 2009; 182: 1694.
EDITORIAL COMMENT The authors present a relatively large series of infants with high grade reflux demonstrating that most could be treated nonoperatively. Indications for surgery were bUTI, worsening VUR grade on 18-month followup VCUG and the appearance of
new renal scars or decreasing relative renal function on DMSA scans. They report efforts to decrease the rate of surgical intervention in this patient group. However, one wonders whether the envelope could be pushed even
DILATING VESICOURETERAL REFLUX CAN BE TREATED NONOPERATIVELY
further. In the absence of infections do these patients require followup VCUG at 18 months? From a renal standpoint is there benefit in correcting persistent VUR? Is bUTI the real game changer? If so, our efforts should be focused on preventing infection by correcting bladder and bowel dysfunction. This would not only decrease the surgical intervention rate from the current 25% but also reduce the morbidity and cost of followup without placing the kidneys at risk. This approach obviously mandates responsible families and effective communication between the urologist and pediatrician. The same argument applies to noninfection related decreasing relative renal function on DMSA scans. The increase in differential function of the unaffected kidney may simply reflect compensatory growth. In the absence of infection
1627
there is no evidence that correcting VUR results in catch-up growth of the affected kidney and reversal of the negative trend in relative renal function. Many questions in VUR can only be answered by objective data from prospective multicenter trials. Only then will we achieve our goal of minimizing the morbidity of treatment and followup while preventing renal injury. Antoine Khoury Department of Urology University of California-Irvine Irvine, California and Pediatric Urology Children’s Hospital of Orange County Orange, California