POST-PYELONEPHRITIC RENAL SCARS ARE NOT ASSOCIATED WITH VESICOURETERAL REFLUX IN CHILDREN

0022-5347/05/1734-1345/0 THE JOURNAL OF UROLOGY® Copyright © 2005 by AMERICAN UROLOGICAL ASSOCIATION

Vol. 173, 1345–1348, April 2005 Printed in U.S.A.

DOI: 10.1097/01.ju.0000152321.16833.f5

POST-PYELONEPHRITIC RENAL SCARS ARE NOT ASSOCIATED WITH VESICOURETERAL REFLUX IN CHILDREN SEPPO TASKINEN*

AND

¨ NNHOLM KAI RO

From the Hospital for Children and Adolescents, Helsinki University, Helsinki, Finland

ABSTRACT

Purpose: Children with pyelonephritis are at risk for renal damage. We assess the value of clinical signs and urological abnormalities in predicting renal scars in children following pyelonephritis. Materials and Methods: A total of 64 hospitalized children (29 females and 35 males, median age 2.9 years) underwent ultrasonography and 99mtechnetium labeled dimercapto-succinic acid (DMSA) scintigraphy imaging within 1 week following the diagnosis of the first pyelonephritis. Voiding cystourethrography was performed 8 weeks after the diagnosis. Followup DMSA scintigraphy was performed in 58 patients after 2 years of followup. Results: Urological abnormalities observed were vesicoureteral reflux (VUR, grade 2 or higher) in 11 patients (19%), nonrefluxing and nonobstructed megaureter in 2 (4%) and pyeloureteral obstruction in 1 (2%). The first DMSA scintigraphy showed parenchymal defects in 48% of patients. VUR did not increase the risk of renal defects. At 2 years after the infection 12 of the 58 patients (21%) had renal scars. Nine of these patients did not have VUR. However, 2 patients with high grade VUR and repeat infections demonstrated deterioration of kidney function during followup. The patients with renal scars were older than those without scars (3.1 vs 0.8 years, p ⫽ 0.0291) at the time of infection. Conclusions: Renal scars after first pyelonephritis are in most cases not associated with abnormalities of the urinary tract, but are caused by the infection itself. However, structural abnormalities may predispose to recurrent infections. Following pyelonephritis new renal scars may develop in all age groups in both sexes. KEY WORDS: technetium tc 99m dimercaptosuccinic acid, pyelonephritis, cicatrix, vesico-ureteral reflux

Symptomatic urinary tract infection (UTI) affects 8.4% of girls and 1.7% of boys by age 7 years.1 About half of the girls and two thirds of boys present with high fever at the time of infection.1 Fever is usually considered the clinical sign of pyelonephritis, although imaging studies often fail to reveal parenchymal involvement.2 About a third of the children with pyelonephritis will have development of renal scars.3 UTIs in children are often associated with urinary tract abnormalities. Typical abnormalities are vesicoureteral reflux (VUR) in 18% to 55% of patients and obstruction in up to 22%.4, 5 The association of urinary tract abnormalities and renal scars is still unclear. The purpose of this study was to assess the value of clinical signs and urological abnormalities in predicting renal scars in children following pyelonephritis. MATERIALS AND METHODS

A total of 64 children who were referred with first onset of pyelonephritis between 1998 and 2000 were included in the study. The patients were primarily hospitalized and treated at the urological or pediatric department. The patients received intravenous antibiotics (100 mg cefuroxime daily) until the fever resolved, followed by oral antibiotics for 7 to 10 days. After that prophylactic medication (2 mg trimethoprim or 1 mg nitrofurantoin daily) was started until voiding cystourethrography (VCUG) was performed. Of the 64 patients 29 were girls and 35 were boys. Age distribution of the girls was 0.1 to 15.5 years (median 1.7) and of the boys was 0.02 to

5 years (0.3). Nine of the 29 girls and 30 of the 35 boys were younger than 1 year at the time of infection. Pyelonephritis was diagnosed clinically with urinalysis showing pyuria (more than 10 white cells per mm3 urine) and bacteriuria (more than 10,000 colony forming units per ml) in 2 samples. In addition, the body temperature had to be 38.5C or greater, or C-reactive protein (CRP) had to be 40 mg/ml or greater (normal less than 10). The patients with unexplained previous fever episodes were excluded. In every patient the clinical symptoms were recorded. During the hospital stay urinary tract ultrasonography (US) and 99mtechnetium labeled dimercapto-succinic acid (DMSA) scintigraphy were performed in all 64 patients using standard methods. For scintigraphy the injection was given 3 hours before imaging. Focal renal defects were considered to be caused by the infection. The split function of the kidneys was considered normal if it was 44% and 56%, or more evenly distributed. VCUG was performed an average of 8 weeks after the infection. In VCUG the bladder was filled until the patient voided or a pressure of 80 to 100 cm contrast material was achieved. Reflux was graded according to the International Reflux Study in Children grading system.6 The results of all imaging studies were confirmed by a pediatric urologist (ST). Laboratory measurements were performed with standard methods. Proteinuria was detected with dipstick. At 2 years DMSA scintigraphy was repeated in 58 patients. The patient with pyeloureteral obstruction (PUO) and the 2 patients with megaureter were excluded from followup. Repeat infections were confirmed with positive laboratory reports. The data with continuous variables were compared using the Mann-Whitney test, and the data with categorical variables were compared using contingency table analysis.

Submitted for publication July 13, 2004. Study received ethical committee approval. * Correspondence: Hospital for Children and Adolescents, University of Helsinki, Stenba¨ckinkatu 11, 00290 Helsinki, Finland (telephone: ⫹358 –50-4272542; FAX: ⫹358 –9-47175314; e-mail: seppo. [email protected]). 1345

1346

POST-PYELONEPHRITIC RENAL SCARS AND REFLUX RESULTS

The maximal fever during the first pyelonephritic attack varied from 38.1C to 41C (median 39.6C). Maximal CRP varied from 5 to 326 mg/l (median 114). Serum creatinine was normal in all patients. The diagnosis of pyelonephritis was made 2 days (range 0 to 9) from the onset of symptoms. In addition to the fever present in all 64 patients, the most common symptoms were vomiting in 10, poor general condition in 7, dysuria in 6, flank pain in 3 and wetting in 1. Urine bacterial culture demonstrated Escherichia coli in 60 cases, Enterococcus faecalis in 2, Enterobacter aerogenes in 1 and klebsiella species in 1. US and VCUG were considered normal in 26 patients. Major abnormalities (VUR, PUO or megaureter) were detected in 14 of the 64 patients (22%), including VUR grade 2 or greater in 11 (17%), nonrefluxing and nonobstructed megaureter in 2 (3%) and PUO in 1 (2%). VUR was evaluated as grade 2 in 3, grade 3 in 6 and grade 4 in 2 patients. Five of the 11 patients with VUR had bilateral reflux. The 3 patients with grade 1 VUR and with no renal scars were included in the nonVUR group. In addition, functional disorders of the lower urinary tract were suspected in 12 of 64 patients (19%) because of enlarged bladder volume (8) and increased residual urine volume or widened posterior urethra (4) in VCUG. Functional disorders were associated in 1 case with VUR and in 1 case with megaureter. Mild nonsignificant dilatation of the renal pelvis or ureter was detected in 20 patients at the time of the initial infection. Excluding patients with PUO and nonrefluxing nonobstructed megaureter, slightly dilated pelvis or ureter was detected in 6 of 11 patients (55%) with VUR and in 14 of 50 (28%) without VUR (p ⫽ 0.2298). The diagnoses of PUO and nonobstructed megaureter were confirmed by excretory urography. DMSA scintigraphy revealed a defect in kidney parenchyma in 31 of 64 children (48%) during the first pyelonephritis (table 1). A total of 11 patients had abnormal split function between the kidneys, of whom 9 also had parenchymal defects. Four of the 11 patients with abnormal split function had VUR, 3 bilaterally. The most reliable clinical signs predicting renal parenchymal defect were high fever (greater than 39C), high CRP (greater than 100 mg/ml) and proteinuria. If all of these conditions were observed, 81% of the patients had renal defects. Presence of VUR, young age, delay in diagnosis or suspicion of bladder dysfunction did not increase the risk of renal defects. In fact, patients with renal defects tended to be older than those without defects (p ⫽ 0.0672). There were no significant differences between the genders. The patients with PUO and nonrefluxing nonobstructed megaureter did not have defects on scintigraphy. At followup DMSA scintigraphy 21% of the patients had permanent renal scars (see figure). In 41% of the patients who had a renal defect the scar developed in the same location as on the initial scintigraphy (table 2). Permanent scars

Age of patients at time of infection and renal scars at followup. Males and females are indicated separately. Box plot indicates medians and 25th and 75th percentiles.

tended to be more common in girls than in boys but the difference was not statistically significant (table 3). The duration of symptoms before diagnosis had no influence on scar development. Permanent scars were associated with high fever at the time of the initial infection but there was no association between renal scars and CRP values or proteinuria. The patients with scars were significantly older at the time of infection than those without scars. In younger patients the defects detected at the time of infection were also healed somewhat better at followup (1.2 ⫾ 1.4 vs 4.7 ⫾ 5.5 years, defect healed or scar developed, p ⫽ 0.0915). At followup DMSA scan only 3 of 58 patients (5%) had abnormal split function. In 2 of these patients who had VUR and recurrent infections the split function of the originally affected kidney decreased from 32% to 16% and from 23% to 19%. Four patients required operative treatment during the 2-year followup period. Pyeloplasty for PUO was performed in 1 patient. One patient underwent ureteral neoimplantation and tapering because of massive VUR and recurrent UTIs. In addition, periureteral collagen injection was performed in 2 patients because of VUR, 1 with recurrent UTIs and 1 with duplex ureters. The patients with VUR or suspected bladder dysfunction continued antimicrobial prophylaxis with trimethoprim (2 to 3 mg/kg) or nitrofurantoin (1 mg/kg). Six of the 58 patients (10%) had new pyelonephritic episodes during the followup period (1 episode in 4, 5 in 1 and 6 in 1). In 4 of these 6 patients repeat infections were associated with VUR. Although there was progression of the renal defects in 2 patients, no new focal defects could be detected in children with repeat infections. DISCUSSION

Vesicoureteral reflux is the most common urological abnormality in patients with pyelonephritis.3, 5 This finding is not unexpected, as reflux allows transport of bacteria from the bladder to the renal pelvis. In this study VUR was detected in only 17% of patients. One explanation for the low incidence of VUR is that sometimes it can be intermittent. Although, as in this study, VCUG is usually done only with 1 bladder filling, it is possible that a second filling increased the VUR rate.7 In addition to VUR, another possible mechanism for ascending infection may be bacterial virulence. Especially P fimbriae and class II pap G allele of E. coli seem to be responsible for pyelonephritis in those patients who do not have VUR.8, 9 VUR has traditionally been detected by VCUG. US has not been sensitive enough to replace VCUG.10 Similarly, in the present study only 55% of patients with VUR had transient mild hydronephrosis on US. At least slight dilatation of the upper urinary tract during the infection was also detected in 28% of patients without VUR or any other structural abnormality. Obstruction of the urinary tract has been detected in less than 10% of patients with urinary tract infection.3 It has also been suggested that imaging of the upper urinary tract is not necessary after the first febrile UTI if there has been a prenatal US screening.5 In this study 1 patient had pyeloureteral obstruction requiring surgery, and 2 patients with nonrefluxing and nonobstructed megaureters were kept in followup with antibiotic prophylaxis. Although fever is a typical clinical sign of pyelonephritis, it has been concluded on the basis of DMSA scintigraphy that only about half of the patients with UTI and fever have parenchymal involvement.11, 12 In the present study DMSA scintigraphy showed focal defects in 48% of patients with a clinical diagnosis of pyelonephritis. Permanent scars developed only in those kidneys that demonstrated defects on DMSA scan at the time of the infection. The risk of renal scars was 21%. In 11 of 64 patients the split function of the kidneys was abnormal during the infection. The abnormal

1347

POST-PYELONEPHRITIC RENAL SCARS AND REFLUX TABLE 1. Characteristics of patients with pyelonephritis and result of DMSA scintigraphy at time of infection Characteristic No. pts No. boys/girls Median age yrs (range) Age less than 1 yr Median max fever (range) Median max B leukocytes ⫻ 109/l (range) Median max CRP mg/l (range) Median days delay in diagnosis (range) No. VUR grade 2 or greater No. nonrefluxing megaureter No. proteinuria No. temperature greater than 39C ⫹ proteinuria No. temperature greater than 39C ⫹ CRP greater than 100 mg/l No. temperature greater than 39C ⫹ proteinuria ⫹ CRP greater than 100 mg/ml

No Defect (25th–75th percentiles)

Defect (25th–75th percentiles)

33 21/12 0.6 (0.1–1.3) 25 39.3C (39.0–39.8) 19.6 (15.5–22.7) 83 (61–144) 3 (1–5) 5 2 13 9 9

31 14/17 0.9 (0.3–4.1) 16 39.9C (39.3–40.0) 18.8 (15.9–18.8) 142 (107–180) 2 (1–4) 6 0 20 18 23

4

17

0.2177 0.0672 0.0424 0.0045 Not significant 0.0008 Not significant — — 0.0785 0.0251 0.0005 0.0007

Girls older than 1 year were especially prone to development of renal scars. However, scars were not common in male infants without massive reflux. This finding is in accordance with previous reports that younger pediatric patients are not at greater risk for renal scarring.13–15 In the present study there may have been some bias in the discovery that girls were more prone to renal scars because only hospitalized patients were included in the study. This finding means that all infants (mostly males) were hospitalized but in the older age group (mostly females) only some of the patients were hospitalized. Thus, we do not know about the development of scars in older girls with outpatient treatment. In this study VUR did not increase the risk of renal defects either in the primary stage or at followup. This finding is similar to those of recent reports showing that VUR is a weak predictor of renal damage after UTI.15, 16 On the other hand, in our study most of the patients with recurrent UTIs had VUR, which has been reported to be a typical risk factor for recurrence.17

TABLE 2. Development of renal scars at followup in 64 pyelonephritis cases indicated on DMSA scintigraphy 2-Yr Followup After Infection (No. VUR) No. with defect at infection: With scar Without scar Lost to followup No. without defect at infection: With scar Without scar Lost to followup

p Value

31 (6) 12 (3) 17 (3) 2 (0) 33 (5) 0 29 (5) 4 (0)

split function remained abnormal in only 3 patients who had severe and probably congenital nephropathic changes. Although the number of renal defects in this study was similar to rates reported previously, it is possible that more sensitive methods like DMSA with single photon emission computerized tomography or magnetic resonance imaging reveal a larger number of even smaller defects. Children with high CRP, high fever and dilating reflux are at greater risk for renal damage than those with a milder clinical picture.4 Our results confirm this finding. The patients with high fever (greater than 39C), high CRP (greater than 100 mg/ml) and proteinuria had an 81% risk of having defects on initial DMSA scintigraphy and a 35% risk of scar formation at followup. In patients without these clinical findings the risk of a defect on initial scintigraphy was 33% and the risk of a scar was 15%. A short delay in the diagnosis of pyelonephritis appears not to be critical for renal scar development.

CONCLUSIONS

In most instances the development of renal scars is not associated with urological abnormalities or with age in children with pyelonephritis. However, VUR can predispose to recurrent UTIs and further renal damage. Although obstructive abnormalities are rare in patients with pyelonephritis, they still exist and should be excluded by US. If we want to find all renal scars, DMSA scintigraphy should be considered in all age groups in both sexes of children with pyelonephritis.

TABLE 3. Pyelonephritic scars at followup and findings at time of pyelonephritis Characteristic No. pts No. boys/girls Median age yrs (range) Median days delay in diagnosis (range) Median max fever (range) No. VUR grade 2 or greater Median max B-leukocytes ⫻ 109/l (range) Median mg/l max CRP (range) No. proteinuria No. temperature greater than 39C ⫹ proteinuria No. temperature greater than 39C ⫹ CRP greater than 100 mg/l No. temperature greater than 39C ⫹ proteinuria ⫹ No. CRP greater than 100 mg/l No. recurrent pyelonephritis

No Scar (25th–75th percentiles)

Scar (25th–75th percentiles)

46 26/20 0.8 (0.1–1.3) 3 (1–5) 39.5C (39.0–39.9) 8 19.6 (15.5–22.5) 111 (77–150) 22 14 21

12 4/8 3.1 (0.5–6.9) 2 (1–3) 40C (39.7–40) 3 16.9 (13.6–24.5) 143 (90–188) 7 6 8

p Value 0.2682 0.0291 Not significant 0.0157 Not significant Not significant 0.1917 Not significant 0.3529 0.3308

11

6

0.158

3

3

0.1804

1348

POST-PYELONEPHRITIC RENAL SCARS AND REFLUX REFERENCES

1. Hellstrom, A., Hanson, E., Hansson, S., Hjalmas, K. and Jodal, U.: Association between urinary symptoms at 7 years old and previous urinary tract infection. Arch Dis Child, 66: 232, 1991 2. Practice parameter: the diagnosis, treatment, and evaluation of the initial urinary tract infection in febrile infants and young children. American Academy of Pediatrics. Committee on Quality Improvement. Subcommittee on Urinary Tract Infection. Pediatrics, 103: 843, 1999 3. Stokland, E., Hellstrom, M., Jacobsson, B., Jodal, U. and Sixt, R.: Renal damage one year after first urinary tract infection: role of dimercaptosuccinic acid scintigraphy. J Pediatr, 129: 815, 1996 4. Spencer, J. R. and Schaeffer, A. J.: Pediatric urinary tract infections. Urol Clin North Am, 13: 661, 1986 5. Hoberman, A., Charron, M., Hickey, R. W., Baskin, M., Kearney, D. H. and Wald, E. R.: Imaging studies after a first febrile urinary tract infection in young children. N Engl J Med, 348: 195, 2003 6. International Reflux Study in Children. International system of radiographic grading of vesicoureteric reflux. Pediatr Radiol, 15: 105, 1985 7. Papandopoulou, F., Efremidis, S. C., Oiconomou, A., Badouraki, M., Panteleli, M., Papachristou, F. et al: Cyclic voiding cystourethrography: is vesicoureteral reflux missed with standard voiding cystourethrography. Eur Radiol, 12: 666, 2002 8. Roberts, J. A.: Pathogenesis of nonobstructive urinary tract infections in children. J Urol, 144: 475, 1990 9. Jantunen, M. E., Siitonen, A., Koskimies, O., Wikstrom, S., Karkkainen, U., Salo, E. et al: Predominance of class II pap G allele of Escherichia coli in pyelonephritis in infants with

normal urinary tract anatomy. J Infect Dis, 181: 1822, 2000 10. Blane, C. E., Di Pietro, M. A., Zerin, J. M., Sedman, A. B. and Bloom, D. A.: Renal sonography is not a reliable screening examination for vesicoureteral reflux. J Urol, 150: 752, 1993 11. Rushton, H. G.: The evaluation of acute pyelonephritis and renal scarring with technetium 99m-dimercaptosuccinic acid renal scintigraphy: evolving concepts and future directions. Pediatr Nephrol, 11: 108, 1997 12. Stokland, E., Hellstrom, M., Jacobsson, B., Jodal, U. and Sixt, R.: Evaluation of DMSA scintigraphy and urography in assessing both acute and permanent renal damage in children. Acta Radiol, 39: 447, 1998 13. Benador, D., Benador, N., Slosman, D., Mermillod, B. and Girardin, E.: Are younger children at highest risk of renal sequelae after pyelonephritis? Lancet, 349: 17, 1997 14. Lin, K. Y., Chiu, N. T., Chen, M. J., Lai, C. H., Huang, J. J., Wang, Y. T. et al: Acute pyelonephritis and sequelae of renal scar in pediatric first febrile urinary tract infection. Pediatr Nephrol, 18: 362, 2003 15. Ditchfield, M. R., Grimwood, K., Cook, D. J., Powell, H. R., Sloane, R., Gulati, S. et al: Persistent renal cortical scintigram defects in children 2 years after urinary tract infection. Pediatr Radiol, 34: 465, 2004 16. Gordon, I., Barkovics, M., Pindoria, S., Cole, T. J. and Woolf, A. S.: Primary vesicoureteric reflux as a predictor of renal damage in children hospitalized with urinary tract infection: a systematic review and meta-analysis. J Am Soc Nephrol, 14: 739, 2003 17. Nuutinen, M. and Uhari, M.: Recurrence and follow-up after urinary tract infection under the age of 1 year. Pediatr Nephrol, 16: 69, 2001