Eur J Pediatr (2012) 171:1605–1610 DOI 10.1007/s00431-012-1775-8

ORIGINAL ARTICLE

Does hydronephrosis predict the presence of severe vesicoureteral reflux? Husam A. Abdulnour & Jonathan L. Williams & John A. Kairalla & Eduardo H. Garin

Received: 9 April 2012 / Accepted: 6 June 2012 / Published online: 27 June 2012 # Springer-Verlag 2012

Abstract We hypothesized that, in patients with vesicoureteral reflux (VUR) grade IV or V, hydronephrosis will likely be found, if the patient has a full bladder during the renal ultrasound examination. Eight hundred thirty-seven patients were included in the study. Patients ranged in age from <1 month to 18.7 years, with a median age of 1.3 years. Five hundred sixty-nine were female and 268 were male. In this retrospective study, each patient underwent a voiding cystourethrogram (VCUG) and a renal ultrasound examination. The presence of hydronephrosis and bladder filling status in 131 renal units with VUR grade IV or V was evaluated. Sensitivity and specificity for hydronephrosis to detect the presence of VUR grades IV and V were 60 and 92 %, respectively. Positive predictive value and negative predictive value were 74 and 87 %, respectively. Odds ratios for the relationship between hydronephrosis and severe VUR was significant (p00.046). Conclusion: In patients with grade IV or V VUR, hydronephrosis will be observed in the presence of a full bladder. Therefore, a renal ultrasound could be considered a screening test to decide on performing a VCUG. H. A. Abdulnour : E. H. Garin (*) Pediatric Nephrology Division, College of Medicine, University of Florida, P.O. Box 100296, Gainesville, FL 32610, USA e-mail: [email protected] J. L. Williams Pediatric Radiology Division, College of Medicine, University of Florida, Gainesville, FL, USA J. A. Kairalla Department of Biostatistics, College of Public Health and Health Professions, University of Florida, Gainesville, FL, USA

Keywords Voiding cystourethrogram . Hydronephrosis . Vesicoureteral reflux . Renal ultrasound

Introduction Vesicoureteral reflux (VUR) is the retrograde flow of urine from the bladder into the ureter and renal pelvis during micturition. This is considered a prominent risk factor for the development of urinary tract infections (UTI) and resulting renal scarring [22, 26, 27, 29]. In infants with a febrile UTI, the Royal College of Physicians recommended the identification of VUR with either a voiding cystourethrogram (VCUG) or radionuclide cystogram (RNC) [4]. However, recently, the National Institute for Health and Clinical Excellence in their 2007 guidelines [18] have suggested VCUG in infants younger than 6 months who present with a recurrent UTI or atypical UTI. Moreover, in 2011, the American Academy of Pediatrics (AAP) limits its recommendations to infants 2 to 24 months of age with a febrile UTI. According to the AAP, a VCUG is indicated if renal and bladder ultrasonography reveals hydronephrosis, scarring, or other findings that would suggest either high-grade VUR or obstructive uropathy and in other atypical or complex clinical circumstances. The AAP committee also suggests obtaining a VCUG if there is a recurrence of a febrile UTI [28]. VCUG and RNC are invasive procedures that require bladder catheterization and expose patients to ionizing radiation. Urethral trauma may occur and there is a slight chance of the patient developing a UTI [8, 15]. In some patients with hydronephrosis, the dilation of the upper urinary tract on ultrasound examination results from severe VUR [1, 13]. It has been suggested that a dilated collecting system in the kidneys on an ultrasound examination

1606

could predict VUR on VCUG. A review of published data on the reliability of hydronephrosis to predict VUR shows conflicting results [3, 5, 7, 9, 14, 21, 30].The variability of the results could be from patient selection (inclusion of patients with different grades of VUR, even those with grade I) and/or technical issues such as patient hydration which determines the urine flow and filling status of the bladder at the time of the renal ultrasound examination. Since urine flows freely between the bladder and the ureter in patients with severe VUR, the purpose of this retrospective study was to evaluate if the presence of hydronephrosis on ultrasound examination could predict VUR grade of IV or V if the patient has a full bladder during the ultrasound examination.

Patients and methods The medical records from pediatric patients with a diagnosis of hydronephrosis, VUR, or UTI (International Classification of Diseases codes), who were followed up at the University of Florida in Gainesville from 1990 to 2009, were reviewed. Patients who underwent VCUG and renal ultrasound examinations were selected, and the presence or absence of VUR and hydronephrosis was documented. The VCUG study was obtained within 2 months of the renal ultrasound. Renal ultrasound and VCUG images of patients with severe (grade IV and V) VUR were then independently reviewed by two authors (HA and JW). In the case of lack of agreement on a specific case, consensus was achieved after the authors (HA, JW, and EHG) reviewed the case together. On renal ultrasound, particular attention was paid to the presence or absence of hydronephrosis and bladder filling status. (a) Hydronephrosis was defined according to the guidelines of the Society for Fetal Urology in those patients whose renal ultrasound was obtained at <1 month of age [19]. For all other patients, hydronephrosis was defined as distention of the renal pelvis and calices. (b) Bladder capacity: Bladder volume was calculated using measurements obtained during bladder ultrasound and compared with published normative data [10]. The volume at the fifth percentile does not represent the fifth percentile volume of a full bladder but the minimum volume associated with the desire to void in the cited study [10]. The fifth percentile bladder volume obtained using the quoted equation is close to the volume for bladder capacity suggested by Koff [11] and Nijman [20]. Therefore, we classified bladder status as full when bladder volume was greater than the fifth percentile for age as per quoted study and empty when bladder walls were collapsed or bladder volume was less than the fifth percentile for age. Since this was a retrospective study, no data about the urge to void during the procedure was available.

Eur J Pediatr (2012) 171:1605–1610

On VCUG, VUR grades were assessed according to the International Grading System for Vesicoureteral Reflux [12]. For the purpose of data analysis, each kidney was considered separately as a “renal unit.” The study was approved by the Institutional Review Board at the University of Florida Gainesville Health Science Center. Statistical analysis Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and estimated odds ratio (with 95 % confidence intervals) were calculated for the tested conditions. Pearson’s chi-square tests were used to test the significance of the odds ratios. Finally, the Breslow–Day test for the homogeneity of the odds ratio was used to compare odds ratios across bladder status and across VUR status. Chi-square test for comparison of ratios was used.

Results Patients One thousand six hundred thirty-two pediatric patient’s medical records were reviewed. Of those, 837 patients had undergone a renal ultrasound and a VCUG. Patients ranged in age from <1 month to 18.7 years, with a median age of 1.3 years. Five hundred sixty-nine were female and 268 were male. A total of 1,667 renal units were available for analysis (Table 1). Nine hundred ninety-four renal units had either no VUR or VUR grade I on VCUG. We included VUR grade I in this group because, by definition, contrast media does not reach the renal pelvis in these patients. Six hundred seventythree renal units had VUR grades II to V, but only one hundred seventy-seven presented VUR grades IV or V. In the group of no VUR or VUR grade I, hydronephrosis was found in 211 renal units on ultrasound (21.2 %). In the group including patients with VUR grades II to V, hydronephrosis was found in 145 renal units (21.5 %) (Table 1). Sensitivity, specificity, PPV, and NPV for hydronephrosis to detect all grades of VUR (grades I–V), VUR grades II to V, VUR grades II to III, and VUR grades IV to V are shown in Table 2. We observed an increase in both sensitivity and specificity the higher the VUR grade included in the evaluation. The table also shows that both the PPV and the NPV increased the higher the VUR grade studied. The estimated odds ratio for the presence of hydronephrosis in non-VUR/VUR grade I group compared to VUR grades II–V group was 1.02, with 95 % confidence interval (CI) of 0.80 to 1.29. The estimated odds ratio for the presence of hydronephrosis in units with VUR grades IV

Eur J Pediatr (2012) 171:1605–1610

1607

Table 1 Presence or absence of VUR in hydronephrotic and nonhydronephrotic renal units

No VUR VUR grades I–V

Subtotal Total

Grade I Grade II Grade III Grade IV Grade V

VUR vesicoureteral reflux

Number of renal units Female Male Hydronephrosis No hydronephrosis

896 567 329 195 701

and V to units with grades II and III was 18.42, with 95 % CI of 11.78 to 28.82. The bladder filling status during ultrasound examination in patients with VUR grades IV and V was reviewed (Fig. 1).There was a total of 199 renal units in 102 patients. Of these 102 patients, 100 had primary VUR. Twenty-one presented with a prenatal diagnosis of hydronephrosis and were studied before 1 month of age. In 79 patients, the diagnosis of VUR was made during the workup for UTI. We include both groups of patients in the analysis because our study was not to evaluate the clinical significance of VUR but to develop guidelines to screen patients for severe VUR. Two patients had posterior urethral valves (PUV). In one, the PUV was discovered at 8 months of age when the patient presented with urosepsis. The other was included because the studies were performed years after the surgical correction of PUV. The renal units of the 102 patients were divided into two groups: One group included renal units with no VUR or VUR grades I to III (n068), and the other group included those units with VUR grades IV and V (n0131). Although there were 177 renal units with grade IV or V VUR (Fig. 1; Table 1), only 131 had available bladder filling status. In the group with no VUR or VUR grades I to III, hydronephrosis was detected in 15 renal units. Of the 15 renal units, a full bladder was observed in five and an empty bladder in ten renal units. In the group including the 131 units with VUR grades IV and V, hydronephrosis was

Table 2 Diagnostic tests performance with regards to VUR and hydronephrosis VUR grade

Sensitivity (%)

Specificity (%)

PPV (%)

NPV (%)

Alla VUR II–V VUR II–III VUR IV–V

21 22 7 60

78 79 39 92

45 41 26 74

53 60 87 87

VUR vesicoureteral reflux, PPV positive predictive value, NPV negative predictive value a

All grades of VUR included

98 77 21 16 82

306 241 65 17 289

190 148 42 21 169

101 56 45 47 54

76 28 48 60 16

771 550 221 161 610

1,667 1,117 550 356 1,311

detected in 78 renal units. Of these, a full bladder was observed in 58 renal units and an empty bladder in 20. Among renal units with VUR grade IV or V, a statistically significant increase in hydronephrosis was observed when the bladder was full compared with when the bladder was empty (p<0.000001). No such statistically significant increase was found between bladder filling status and hydronephrosis in renal units with no VUR or VUR grades I to III (p00.280). Among renal units with VUR grades IV and V, the estimated odds ratio for the presence of hydronephrosis with a full bladder to the presence of hydronephrosis with an empty bladder was 4.05, with 95 % CI of 1.18 to 5.14. In contrast, among renal units with no VUR or VUR grades I, II, and III, the estimated odds ratio for the presence of hydronephrosis with a full bladder to the presence of hydronephrosis with an empty bladder was 0.62, with 95 % CI of 0.19 to 1.98, which was not statistically significant. The Breslow–Day test for the homogeneity of the odds ratios supports the hypothesis that odds ratios differ by VUR grade, with more severe VUR associated with a stronger hydronephrosis/full bladder relationship (p00.045). Among renal units with a full bladder, the estimated odds ratio for VUR grade IV or V in the presence of hydronephrosis to VUR grade IV or V in the absence of hydronephrosis was highly significant at 9.06, with 95 % CI of 3.72 to 22.08. In contrast, among renal units with an empty bladder, the estimated odds ratio for VUR grade IV or V in the presence of hydronephrosis to VUR grade IV or V in the absence of hydronephrosis was 2.28, with 95 % CI of 0.8 to 6.47, which was not statistically significant. The Breslow– Day test for the homogeneity of the odds ratio supports the hypothesis that odds ratios differ by bladder status, with a full bladder associated with a stronger hydronephrosis and severe VUR relationship (p00.046).

Discussion There are several imaging techniques available to detect VUR, including VCUG, direct and indirect RNC, and cystosonography. The most commonly used technique is

1608

Eur J Pediatr (2012) 171:1605–1610

Fig. 1 Flow diagram showing main study findings. Grade of reflux, presence or absence of hydronephrosis and filling bladder status

Our data showed that, when all grades of VUR are evaluated, the sensitivity and specificity for hydronephrosis to predict VUR are 21 and 78 %, respectively, comparable with other published findings. When we evaluated the data in this study including only patients with VUR grades IV and V, the sensitivity and specificity increased to 60 and 92 %, respectively. Furthermore, when we incorporate the bladder filling status in the evaluation of our data, the estimated odds ratio for VUR grade IV or V in the presence of hydronephrosis to VUR grade IV or V in the absence of hydronephrosis was highly significant. The low sensitivity previously reported could be attributed to the inclusion of all VUR grades in the analysis and the absence of bladder filling status as an item in the data. This conclusion is supported by the analysis of the PPV for renal ultrasound showing hydronephrosis to detect the presence of any VUR grade. In our study, the PPV for all types of VUR grades was 45 %, similar to those reported by Oostenbrink et al. [21] (PPV, 51 %) but higher than Zamir et al. [30] (PPV, 24 %). In our series, the proportion of VUR grades I and II was 52 %, while in the studies of Oostenbrink

fluoroscopic VCUG requiring bladder catheterization and exposure of patients to ionizing radiation. Because only 35 % of unselected patients with a UTI present with VUR [2, 24], there is the need for a noninvasive test prior to performing a VCUG to increase the yield of finding VUR on a VCUG and eliminate ionizing radiation exposure to normal subjects. The presence of hydronephrosis has been suggested as a screening test to predict the presence of VUR. A review of published reports [3, 5, 7, 9, 14, 21, 30] gives conflicting results as to the sensitivity and specificity for hydronephrosis on a renal ultrasound examination to screen for the presence of VUR (Table 3). In all quoted studies [3, 5, 7, 9, 14, 21, 30], VUR grades I–V were not analyzed separately and bladder filling was not considered. In this report, our findings indicate that the presence of hydronephrosis does help to screen for VUR grades IV and V if the bladder is full at the time of the ultrasound examination. In addition, we will suggest why previous studies have shown conflicting results.

Table 3 Sensitivity, specificity, PPV†, and NPV‡ in published data and in our data Author

Number of patients

Number of renal units

VUR

Sensitivity (%)

Specificity (%)

PPV (%)

NPV (%)

Berrocal 2007a Oostenbrink 2000b Mahant 2002a Davey 1997a Goldman 2000a Zamir 2004b Hoberman 2003a

573 140 162 455 45 255 302

1,146

All All All All All All All

64 57 40 21 40 18 10

36 81 76 69 33 88 90

26 51 32 7 27 24 40

74 83 82 89 92 83 61

This study

837

22

79

41

60

910 88

1,667

grades grades grades grades grades grades grades

Grades II–V

VUR vesicoureteral reflux, PPV positive predictive value, NPV negative predictive value a

Values were calculated by the authors based on data available from authors

b

Published values

Eur J Pediatr (2012) 171:1605–1610

and Zamir, the PPV was 24 and 69 %, respectively. The correlation coefficient between increased proportion of VUR grades I to III and decreased PPV was −0.9, indicating a strong indirect correlation. The higher the number of patients with VUR grades I–III included in any study, the lower the PPV value. To our knowledge, this is the first study that analyzed the PPV of hydronephrosis and VUR grade. In our study group, if we include only patients with VUR grade IV or V in the analysis, we found a PPV of 74 %. When NPV data were analyzed according to the VUR grade, similar observations are identified. For our data, the NPV for all VUR grades was 53 %, lower than the NPV given by Oostenbrink [21] (84 %) and Zamir [30] (83 %). In this study, if only VUR grades IV and V are evaluated, the NPV improved to 87 %. The diagnostic odds ratio evaluates how much more likely the presence of hydronephrosis is indicative of VUR than the absence of hydronephrosis. Our analysis reveals that, among renal units with a full bladder on ultrasound, the estimated odds ratio for VUR grade IV or V in the presence of hydronephrosis to VUR grade IV or V in the absence of hydronephrosis was highly significant. This supports our hypothesis that hydronephrosis is detected in patients with severe VUR in the presence of a full bladder. The significance of VUR as a predisposing factor for UTI is being reassessed. Recent studies have reported that mild to moderate VUR (grades I to III) do not predispose to UTI, pyelonephritis, or renal scarring [6, 16, 17, 23, 25, 28]. Other published data suggest that there is no difference in the incidence of recurrent UTI in the presence or in the absence of VUR grades I–III. It is generally accepted that VUR grades IVand V predispose to acute pyelonephritis and renal scarring, although no controlled studies have been done to confirm this clinical impression [26]. Until these issues are resolved, it is necessary to investigate patients with acute pyelonephritis for the presence of VUR grade IV or V. This study indicates that, in patients with VUR grade IV or V, hydronephrosis will likely be found on a renal ultrasound examination in the presence of a full bladder. Some patients with acute pyelonephritis may present with hydronephrosis from ureterovesical junction or ureteropelvic junction obstruction and may not have VUR. However, the benefit of finding VUR grades IV and V in patients with a febrile UTI and hydronephrosis may outweigh the risk of a VCUG. We suggest that, in those patients who present with the clinical or imaging findings of acute pyelonephritis, a renal ultrasound examination with a full bladder should be obtained. Fluids should be administered until the desire to void is present in the older child and, in the younger child, until bladder volume achieves the standard volumes for full bladder according to age.

1609

If hydronephrosis is found, we suggest proceeding with a VCUG to confirm the presence of VUR grade IV or V. In the absence of hydronephrosis, the patient may still have VUR grades I to III. Because the recent clinical data of these VUR grades do not support a role in predisposing to recurrent UTI and/or acute pyelonephritis, there may be no need to confirm the presence of VUR in these cases; the VCUG may be deferred. This approach will reduce the number of VCUG requested, therefore minimizing the trauma of the procedure, the irradiation to the patient, and the added cost of the patient evaluation.

Conflict of interest No organization sponsored this research. Therefore, the authors report no conflict of interest.

References 1. Aboutaleb H, Bolduc S, Bagli DJ, Khoury AE (2003) Correlation of vesicoureteral reflux with degree of hydronephrosis and the impact of antireflux surgery. J Urol 170:1560–62 2. Arant BS Jr (1992) Medical management of mild and moderate vesicoureteral reflux: followup studies of infants and young children. A preliminary report of the Southwest Pediatric Nephrology Study Group. J Urol 148:1683–7 3. Berrocal T, Pinilla I, Gutierrez J, Prieto C, de Pablo L, Del Hoyo ML (2007) Mild hydronephrosis in newborns and infants: can ultrasound predict the presence of vesicoureteral reflux. Pediatr Nephrol 22:91–6 4. Chantler C, Berman LH, Colleen-Jones F, Gordon I, Grunberg RN, haycock GB, Hopkins A, Maskell M, Peter LH, Ramsley PG, Sherwood T, Smellie JM, Smyth RL, Valman HB, Verrier-Jones K, Whitaker RH, White RHR (1991) Guidelines for the management of acute urinary tract infection in childhood. Report of a Working Group of the Research Unit, Royal College of Physicians. J R Coll Physicians Lond 25:36-42 5. Davey MS, Zerin JM, Reilly C, Ambrosius WT (1977) Mild renal pelvic dilatation is not predictive of vesicoureteral reflux in children. Pediatr Radiol 27:908–11 6. Garin EH, Olavarria F, Garcia Nieto V, Valenciano B, Campos A, Young L (2006) Clinical significance of primary vesicoureteral reflux and urinary antibiotic prophylaxis after acute pyelonephritis: a multicenter, randomized, controlled study. Pediatrics 117:626–32 7. Goldman M, Lahat E, Strauss S, Reisler G, Livne A, Gordin L, Aladjem M (2000) Imaging after urinary tract infection in male neonates. Pediatrics 105:1232–5 8. Guignard JP (1979) Urinary infection after micturating cystography. Lancet 1:103 9. Hoberman A, Charron M, Hickey, Baskin M, Kearney DH, Wald ER (2000) Imaging studies after a first febrile urinary tract infection in young children. N Engl J Med 348:195–202 10. Kaefer M, Zurakowski D, Bauer SB, Retik AB, Peters CA, Atala A, Treves ST (1997) Estimating normal bladder capacity in children. J Urol 158:2261–2264 11. Koff SA (1983) Estimating bladder capacity in children. Urology 129:552–555 12. Lebowitz RL, Olbing H, Parkkulainen KV (1985) International system of radiographic grading of vesicoureteric reflux. International Reflux Study in Children. Pediatr Radiol 15:105–9

1610 13. Lyon RP, Marshall S, Tanagho EA (1969) The ureteral orifice: its configuration and competency. J Urol 102:504–9 14. Mahant S, Friedman J, MacArthur C (2002) Renal ultrasound findings and vesicoureteral reflux in children hospitalized with urinary tract infection. Arch Dis Child 86:419–422 15. Maskell R, Pead L, Vinnicombe J (1978) Urinary infection after micturating cystography. Lancet 2:1191–2 16. Montini G, Rigon L, Zucchetta P, Fregonese F, Toffolo A, Gobber D, Cecchin D, Pavanello L, Molinari PP, Maschio F, Zanchetta S, Cassar W, Casadio L, Crivellaro C, Fortunati P, Corsini A, Calderan A, Comacchio S, Tommasi L, Hewitt IK, Da Dalt L, Zacchello G, Dall'Amico R, IRIS Group (2008) Prophylaxis after first febrile urinary tract infection in children? A multicenter, randomized, controlled, noninferiority trial. Pediatrics 122:1064–71 17. Montini G, Hewitt I (2009) Urinary tract infections: to prophylaxis or not to prophylaxis? Pediatr Nephrol 24:1605–9 18. National Institute for Health and Clinical Excellence (2010) NICE clinical guideline 54; urinary tract infection in children: diagnosis, treatment and long term management. Available at http://www.nice.org.uk 19. Nguyen HT, Herndon CD, Cooper C, Gatti J, Kirsch A, Kokorowski P, Lee R, Perez-Brayfield M, Metcalfe P, Yerkes E, Cendron M, Campbell JB (2010) The Society for Fetal Urology consensus statement on the evaluation and management of antenatal hydronephrosis. J Pediatr Urol 6:212–31 20. Nijman RJM (2009) Urodynamic studies of the lower urinary tract. In: Gearhart JG, Rink RC, Mouriquand PDE (eds) Pediatric urology: expert consult, 2nd edn. Saunders Elsevier, Philadelphia, pp 355–356 21. Oostenbrink R, van der Heijden AJ, Moons KG, Moll HA (2000) Prediction of vesico-ureteric reflux in childhood urinary tract infection: a multivariate approach. Acta Paediatr 89:806–10 22. Orellana P, Baquedano P, Rangarajan V, Zhao JH, Eng ND, Fettich J, Chaiwatanarat T, Sonmezoglu K, Kumar D, Park YH, Samuel AM, Sixt R, Bhatnagar V, Padhy AK (2004) Relationship between

Eur J Pediatr (2012) 171:1605–1610

23.

24. 25.

26.

27.

28.

29.

30.

acute pyelonephritis, renal scarring, and vesicoureteral reflux. Results of a coordinated research project. Pediatr Nephrol 19:1122–6 Oualha M, Larakeb SA, Dunand O, Ulinski T, Bensman A (2009) Stop of antibiotic prophylaxis in mild/moderate grade vesicoureteral reflux. Arch Pediatr 17:3–9 Rolleston GL, Shannon FT, Utley WL (1975) Follow-up of vesicoureteric reflux in the newborn. Kidney Int Suppl 4:S59–64 Roussey-Kesler G, Gadjos V, Idres N, Horen B, Ichay L, Leclair MD, Raymond F, Grellier A, Hazart I, de Parscau L, Salomon R, Champion G, Leroy V, Guigonis V, Siret D, Palcoux JB, Taque S, Lemoigne A, Nguyen JM, Guyot C (2008) 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 179:674–9 Shaikh N, Ewing AL, Bhatnagar S, Hoberman A (2004) Risk of renal scarring in children with a first urinary tract infection: a systematic review. Pediatrics 126:1084–91 Soylu A, Demir BK, Turkmen M, Bekem O, Saygi M, Cakmakçi H, Kavukçu S (2008) Predictors of renal scar in children with urinary infection and vesicoureteral reflux. Pediatr Nephrol 23:2227–32 Subcommitte on Urinary Tract Infection, Steering Committee on Quality Improvement and Management (2011) Urinary tract infection: clinical practice guideline for the diagnosis and management of the initial UTI in febrile infants and children 2 to 24 months. Pediatrics 128:595–604 Supavekin S, Kucivilize K, Hunnangkul S, Sriprapaporn J, Pattaragarn A, Sumboonnanonda A (2006) The relation of vesicoureteral reflux and renal scarring in childhood urinary tract infection. J Med Assoc Thai 89(Suppl 2):S41–7 Zamir G, Sakran W, Horowitz Y, Koren A, Miron D (2004) Urinary tract infection: is there a need for routine renal ultrasonography? Arch Dis Child 89:466–8