- Email: [email protected]
IMPACT OF VESICOURETERAL REFLUX ON THE SIZE OF RENAL LESIONS AFTER AN EPISODE OF ACUTE PYELONEPHRITIS
0022-5347/05/1732-0571/0 THE JOURNAL OF UROLOGY® Copyright © 2005 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 173, 571–575, February 2005 Printed in U.S.A.
DOI: 10.1097/01.ju.0000151263.36909.91
IMPACT OF VESICOURETERAL REFLUX ON THE SIZE OF RENAL LESIONS AFTER AN EPISODE OF ACUTE PYELONEPHRITIS ´ LEZ,* JEAN-PIERRE PAPAZYAN ELSA GONZA
AND
ERIC GIRARDIN
From the Paediatric Nephrology Unit, Department of Paediatrics, Children’s Hospital (EG, EG) and Division of Nuclear Medicine, Department of Radiology, Geneva University Hospital (J-PP), Geneva, Switzerland
ABSTRACT
Purpose: We determined the impact of vesicoureteral reflux (VUR) on the size of renal lesions in children after an episode of acute pyelonephritis. Materials and Methods: A total of 161 children (mean age 2.44 years) with acute pyelonephritis were studied. All had renal lesions on dimercapto-succinic acid scintigraphy done at admission to the hospital. A second dimercapto-succinic acid scan was performed at 3 months. Voiding cystourethrography was done at 6 weeks and VUR was graded I to V. For each renal unit layouts of renal lesions were drawn, and the damage surface was calculated and reported for the total surface of the kidney. Results: Mean size of acute lesions and scars increased with severity of reflux (p ⬍0.0001), with an important overlap of individual values. Mean size of renal scars in the group of renal units with acute lesions was 5.8% ⫾ 8.5% in patients without VUR, 9.9% ⫾ 7.3% in those with grade I reflux, 7.7% ⫾ 11.0% in those with grade II reflux, 17.7% ⫾ 14.7% in those with grade III reflux and 17.4% ⫾ 27.7% in those with grade IV reflux (p ⬍0.001). The size of renal lesions decreased significantly with time. The rate of regression of lesions decreased with increasing reflux. When analyzed according to 3 age groups sizes of scars increased significantly with age. Conclusions: VUR has an impact on the size of renal lesions after an episode of pyelonephritis. Children with a grade III or IV reflux are more likely to have larger renal scars. On the other hand, acute lesions of important size may develop even in the absence of VUR. KEY WORDS: pyelonephritis, cicatrix, vesico-ureteral reflux, technetium Tc 99m dimercaptosuccinic acid
Acute pyelonephritis in children often leads to irreversible lesions of renal parenchyma that are still present several months after the acute episode and defined as renal scars.1 It has been shown that the inflammatory process, which appears early in the course of the disease, is responsible for scar formation.2 Children with renal lesions after repeated episodes of acute pyelonephritis are at risk for development of hypertension and impairment of renal function.3–5 In pediatric pyelonephritis acute lesions of the kidney on dimercapto-succinic acid (DMSA) scan are observed in 42% to 92% of patients, leading to permanent renal damage in 36% to 68%.6 –10 The gold standard for the diagnosis of these lesions is DMSA scintigraphy, which has good sensitivity and specificity.6, 11 Vesicoureteral reflux (VUR) is classically considered a risk factor for the development of renal scarring but the development of scars has been observed to occur in the absence of VUR.6 –9 This finding raises the question of the exact correlation between VUR and renal lesions, and of the importance of VUR in the etiology of renal scarring. In the literature there is a lack of information concerning the size of renal lesions on DMSA scan. Several studies demonstrate DMSA results as positive or negative. A positive DMSA scan may correspond to small lesions of limited clinical significance. Thus, it is important to measure the size of renal lesions on DMSA scan, especially if one wants to correlate renal damage with potential risk factors such as VUR. The aim of this study was to determine the impact of VUR on the size of acute renal lesions and resultant renal scars. For this purpose the frequency of renal lesions and their
precise size on DMSA scintigraphy were analyzed during acute episodes and at 3 months. METHODS
Patients with clinically suspected pyelonephritis had a urine sample taken for culture and a blood sample for blood count, C-reactive protein determination and culture. Clinical suspicion was defined as fever greater than 38.5C, general signs of feeding difficulties, vomiting, irritability and abdominal or flank pain in older children. If the urine culture was positive, renal ultrasound and DMSA renal scan were performed in the acute phase (3 to 5 days). All patients presented initially between 1994 and 1997. Patients were included in the study if they had a positive urine culture and an acute lesion on the first DMSA renal scintigraphy. A positive urine culture was defined as 100,000 colonies per ml or greater with 1 type of bacteria (2 if the bacteria were Escherichia coli and Enterococcus faecalis) in a urine sample collected in a sterile bag, or any colonies if the sample was collected by suprapubic puncture of the bladder. Patients were excluded if they had obstructive renal disease, incomplete radiological files or a scintigram of less than 1 ⫻ 0.8 inches, which does not allow precise measurement of the surface of the lesions. Renal scintigraphy was performed according to standard techniques—3 to 4 hours after an intravenous injection of DMSA labeled 3.7 MBq 99mtechnetium per kilogram (with a minimum dose of 18.5 MBq and a maximum dose of 185 MBq) image acquisition was performed with a GCA9300A/HG gamma camera connected to a computer (Toshiba Medical Systems Co., Japan). Six different views were obtained (1 posterior, 2 posterior oblique, 1 anterior and 2 anterior oblique projections). Acute lesions were defined as a
Submitted for publication April 23, 2004. * Correspondence: Department of Paediatrics, Children’s Hospital, 6 rue Willy-Donze´, 1211 Gene`ve 14, Switzerland (e-mail: gonzaea2@ hotmail.com). 571
572
IMPACT OF VESICOURETERAL REFLUX ON RENAL LESIONS AFTER PYELONEPHRITIS
decreased uptake of isotope in at least 2 different views. Scars were defined as persistent changes in the same location as in the first DMSA scan. Every image was independently assessed by 2 experienced radiologists who used standard criteria previously defined by Patel et al.12 To calculate the precise size of renal lesions, each image was assessed regardless of clinical history and presence or absence of vesicoureteral reflux. Each outline of lesion and kidney was drawn on the view of the scintigram where the lesion was the largest. If multiple lesions were present, surface areas were added together. The corresponding surfaces were calculated and compared to the total surface of the kidney. Sizes of lesions were expressed as percentage of total kidney surface. To diagnose VUR, voiding cystourethrography (VCUG) was done 6 weeks after the diagnosis. Reflux was classified according to the gradation of the International Reflux Study committee, grade I to V.13 The surfaces of renal lesions were correlated with the result of the VCUG. Size of renal lesions was analyzed on first and second DMSA scan. To study the relation between clinical characteristics, grade of reflux and renal lesions, in case of bilateral reflux and/or lesion the side where the reflux and/or lesion was greatest was considered. Acute renal lesions and scars were further analyzed according to 4 groups of patients—those with left dominant reflux (defined as left unilateral reflux or bilateral reflux with a more important grade on the left side), those with right dominant reflux (defined as right unilateral reflux or bilateral reflux with a more important grade on the right side), those with bilateral reflux of the same grade and those without reflux. Finally, renal units with acute lesions were analyzed according to 3 age groups—less than 1 year, 1 to 5 years and more than 5 years. Results were given in terms of renal units as mean ⫾ standard error. Differences between percentages were analyzed using the chi-square test. Renal lesion sizes were compared using the Kruskal-Wallis nonparametric test and Dunn’s posttest. Differences between means were analyzed using the Mann-Whitney U test and Wilcoxon rank sum test for paired data. The effect of reflux on the progression of lesions and the effect of clinical characteristics on the size of scars were analyzed using analysis of variance. Statistical analysis was performed using the Stata statistical software package (StataCorp LP, College Station, Texas) and Prism software (GraphPad Software, Inc., San Diego, California). RESULTS
Characteristics of study population. A total of 161 children (age range birth to 11.2 years, 120 girls and 41 boys) were included in this study. Of the boys 78% were younger than 1 year at the time of diagnosis. Mean age was 1.17 years (median 0.47) for boys and 2.87 years (1.82) for girls. Clinical data are outlined in table 1. The impact of the clinical parameters on the size of renal scars was studied using analysis of variance. The presence of a previous urinary tract infection (UTI, 12 patients) and abnormal voiding habits (9) were found to be significantly related to the size of renal scars (p ⬍0.001 and p ⫽ 0.04, respectively). Mean size of renal scars in the group of patients with previous UTI was 19.1% ⫾ 20.6% compared to 8.1% ⫾ 10.5% in the other patients (p ⫽ 0.038).
TABLE 1. Characteristics of study population
Frequency of renal lesions and VUR. According to the inclusion criteria, each child had a renal lesion on the first DMSA scan. In terms of renal units, 197 of 322 renal units (61%) had lesions on the first DMSA scan. On the second DMSA scan performed at 3 months 80% of renal lesions exhibited partial or total regression, and a renal scar was present in 104 of 322 renal units (32%). Of 322 renal units vesicoureteral reflux was diagnosed in 89 (28%) on VCUG at 6 weeks. Grades of reflux were distributed as follows—15 renal units (17%) grade I, 38 (43%) grade II, 26 (29%) grade III and 10 (11%) grade IV. Among renal units with acute lesions 47% in the group without reflux had renal scars, 60% had scars in the group with reflux grades I and II, and 78% had scars in the group with reflux grades III and IV. The frequency of renal scars increased with the severity of reflux (p ⫽ 0.003). Size of renal lesions and VUR. When analyzed according to the grade of reflux an overlap of individual values was observed. Nevertheless, the mean size of acute renal lesions increased when reflux was present (Mann-Whitney U test between renal units with and without reflux, p ⬍0.0001, fig. 1, A). Of the subgroup of renal units with acute lesions the mean size of renal lesions was 19.1% ⫾ 12.1% in the group without VUR, 25.2% ⫾ 8.8% in the group with grade I VUR, 18.0% ⫾ 10.4% in the group with grade II VUR, 24.0% ⫾ 12.4% in the group with grade III VUR and 27.6% ⫾ 25.4% in the group with grade IV VUR. Overlap was also observed for renal scars. Mean size increased with severity of reflux (Kruskal-Wallis nonparametric test, p ⬍0.0001, fig. 1, B). Of the subgroup of renal units with acute lesions the mean size of renal scars was 5.8% ⫾ 8.5% in the group without VUR, 9.9% ⫾ 7.3% in the group with grade I VUR, 7.7% ⫾ 11.0% in the group with grade II VUR, 17.7% ⫾ 14.7% in the group with grade III VUR and 17.4% ⫾ 27.7% in the group with grade IV VUR (KruskalWallis nonparametric test, p ⬍0.001). The risk for a child having a renal scar of more than 25% of the surface of the kidney increased with grade of reflux (p ⫽ 0.015, table 2). When analyzing regression of lesions at 3 months in the group of renal units with an acute lesion the size of renal lesions decreased significantly with time in all groups. The grade of reflux affected significantly the rate of regression of lesions. The greater the reflux, the lesser the regression (ANOVA, p ⬍0.001, table 3). To highlight the correlation between VUR and renal lesions for a given patient, renal lesions were analyzed by subtracting the size of lesions of 1 side from the other (right minus left) for each patient and analyzed according to 4 groups—patients with left dominant reflux, right dominant reflux, reflux of same grade and without reflux. In patients with left dominant reflux the mean size of acute renal lesions and scars on the left side was significantly larger than in those with right dominant reflux (p ⬍0.05 and p ⬍0.01, respectively), indicating that renal lesions were more extensive when reflux was present on the same side. Size of renal lesions and age. The frequency and size of renal scars were analyzed according to 3 age groups— younger than 1 year, 1 to 5 years and older than 5 years. The frequency of renal scarring increased with age (p ⫽ 0.0003). When considering only renal units with an acute lesion sizes of scars increased significantly with age (p ⬍0.0001), and posttest revealed statistically significant differences between the group of patients younger than 1 year and the other 2 age groups (p ⬍0.001, fig. 2).
No. Pts Abnormal voiding habits 9 Abnormal bowel habits 7 Previous UTI 12 Abnormal urethra on VCUG 6 Abnormal bladder on VCUG* 5 * One patient had bladder thickening and 4 had bladder diverticuli.
DISCUSSION
Renal lesions may lead to hypertension and renal function impairment. Thus, one of the main goals of pyelonephritis management is to decrease the occurrence of renal scars. As reflux has traditionally been considered a risk factor for renal
IMPACT OF VESICOURETERAL REFLUX ON RENAL LESIONS AFTER PYELONEPHRITIS
573
FIG. 2. Size of renal scars in group of renal units with acute lesions according to age—younger than 1 year, 1 to 5 years and greater than 5 years (p ⬍0.001).
FIG. 1. Size of renal lesions on DMSA scintigraphy according to grade of reflux. A, acute lesions are demonstrated on DMSA scan. B, renal scars are shown on DMSA scan performed at 3 months. TABLE 2. Risk of scar more than 25% of total surface of kidney according to reflux grade No. pts without reflux/total No. (%) No. pts with grades I ⫹ II reflux (%) No. pts with grades III ⫹ IV reflux (%)
3/50 (6) 3/22 (14) 8/20 (40)
TABLE 3. Regression of renal lesions related to reflux grade % Acute Lesions
% Scars
p Value
No reflux 19.1 ⫾ 12.1 5.7 ⫾ 8.4 ⬍0.0001 Grades I and II reflux 20.1 ⫾ 10.4 8.3 ⫾ 10.1 ⬍0.0001 Grades III and IV reflux 24.0 ⫾ 16.2 18.0 ⫾ 17.6 0.0051 Analysis of variance showed that time (p ⬍0.001) and degree of reflux (p ⬍0.001) influenced size of lesions.
lesions,14, 15 considerable efforts were undertaken to diagnose and correct VUR. Nevertheless, several studies indicate that scars are fre-
quently observed in the absence of reflux.6, 8, 16 This observation leads to the question of the exact correlation between reflux and renal lesions. The answer has a direct impact on the clinical management of acute pyelonephritis and on radiological investigations to be performed after an episode of pyelonephritis. DMSA scintigraphy is considered the gold standard for detecting renal lesions. Several studies show DMSA results as positive or negative. Thus, a positive DMSA scan may correspond to small lesions of limited clinical significance. To predict the outcome of acute lesions, several authors have classified the abnormalities demonstrated on the DMSA scan. Goldraich and Goldraich proposed a 4-grade classification depending on the number of lesions, the persistence of normal parenchyma between lesions and the atrophic aspect of the kidney.11 Risdon et al observed that with a classification of 3 types the majority of type C lesions, corresponding to focal defects with complete loss of uptake and loss of renal outline, persisted as irreversible scarring.17 Jakobsson et al quantified lesions on DMSA scan and found that reflux of grade III or more was associated with significantly greater defects (defined as more than 10% of the kidney with DMSA uptake less than 2 SD of the control kidney).8 In the literature there is a lack of information concerning the exact size of renal lesions on DMSA scan. In the present study renal lesions were measured in terms of percentage of the area of the kidney and correlated with the grade of reflux determined by DMSA scintigraphy. Each child enrolled in this study had a renal lesion on DMSA scan done at the time of diagnosis, and 56% had renal scars on the second scintigraphy performed at 3 months. In terms of renal units acute lesions were present in 61% initially and persisted in 53% at 3 months. At the time of the design of the study 3-month followup was considered appropriate to diagnose scars. It was subsequently revealed that some lesions may further regress by 5 months.10 Nevertheless, the frequencies observed in our study are in the range of results previously reported.6, 8, 10 Are the renal lesions observed a consequence of infection or a consequence of preexisting congenital damage? It is difficult to evaluate the origin of lesions on the basis of DMSA scintigraphy alone.18 Clues exist to indicate that congenital
574
IMPACT OF VESICOURETERAL REFLUX ON RENAL LESIONS AFTER PYELONEPHRITIS
lesions did not represent a major cause of renal lesions observed in our study population. Patients with an obstructive disease were excluded, and none of the patients included in the study had hydronephrosis or dysplasic kidneys on fetal ultrasound. Moreover, regression of lesions was observed in 80% of the cases, indicating that at least some of the lesions were related to acute inflammation. Our data confirmed the role of VUR as a risk factor for renal lesions. Mean size of renal scars increased significantly with grade of reflux, and children with grade III to IV reflux were more likely to have development of a larger scar than those with grades I and II. When data were analyzed for each individual patient sizes of renal lesions were larger when ipsilateral reflux was present. These results might underestimate the importance of VUR, as data were collected after a first episode of acute pyelonephritis for the majority of patients, and, thus, do not give information on the long-term impact of VUR on renal lesions. In our population voiding habits and a history of UTI were additional risk factors for larger scars. Longitudinal studies are needed to evaluate this impact. It has been suggested that DMSA scintigraphy might be performed only in patients in whom VCUG shows VUR.19 Our data demonstrate that when individual renal lesions were correlated with grade of reflux value overlaps did not allow prediction of size of acute renal lesions or renal scars according to the absence or presence of reflux and to its grade, nor prediction of the degree of reflux from the size of the renal lesions (fig. 1). Moreover, large acute renal lesions and scars also occurred in the absence of VUR. To evaluate the effect of age on the risk of developing scars, we analyzed our data according to 3 age groups—younger than 1 year, 1 to 5 years and older than 5 years. Previous studies have revealed controversial data concerning renal scarring and age. It is generally admitted that infants are at higher risk for renal scarring than older children but in our study the mean size of renal lesions increased with age. These results confirmed data from recent studies showing that older children are at higher risk for persistent renal lesions.10, 20 Moreover, the relationship between reflux and size of renal scars was similar when considering only children older than 1 year (data not shown). CONCLUSIONS
These data demonstrate that VUR has an impact on the size of renal lesions after an episode of pyelonephritis, and that children with grade III or IV reflux are more likely to have development of larger renal scars. On the other hand, acute lesions of important size may develop even in the absence of VUR. In the treatment of acute pyelonephritis in children, the choice of the radiological investigation to be performed, DMSA scintigraphy and/or VCUG, should be made knowing that one investigation cannot predict the result of the other. Bernadette Mermillod provided advice on statistical analysis. REFERENCES
1. Jakobsson, B., Berg, U. and Svensson, L.: Renal scarring after acute pyelonephritis. Arch Dis Child, 70: 111, 1994 2. Bille, J. and Glauser, M. P.: Protection against chronic pyelonephritis in rats by suppression of acute suppuration: effect of colchicine and neutropenia. J Infect Dis, 146: 220, 1982 3. Jacobson, S. H., Eklof, O., Eriksson, C. G., Lins, L. E., Tidgren, B. and Winberg, J.: Development of hypertension and uraemia after pyelonephritis in childhood: 27 year follow up. BMJ, 299: 703, 1989 4. Smellie, J. M., Prescod, N. P., Shaw, P. J., Risdon, R. A., Bryant, T. N.: Childhood reflux and urinary infection: a follow-up of 10 – 41 years in 226 adults. Pediatr Nephrol, 12: 727, 1998
5. Patzer, L., Seeman, T., Luck, C., Wuhl, E., Janda, J. and Misselwitz, J.: Day- and night-time blood pressure elevation in children with higher grades of renal scarring. J Pediatr, 142: 117, 2003 6. Majd, M. and Rushton, H. G.: Renal cortical scintigraphy in the diagnosis of acute pyelonephritis. Semin Nucl Med, 22: 98, 1992 7. Benador, D., Benador, N., Slosman, D. O., Nussle, D., Mermillod, B. and Girardin, E.: Cortical scintigraphy in the evaluation of renal parenchymal changes in children with pyelonephritis. J Pediatr, 124: 17, 1994 8. Jakobsson, B., Nolstedt, L., Svensson, L., Soderlundh, S. and Berg, U.: 99mTechnetium-dimercaptosuccinic acid scan in the diagnosis of acute pyelonephritis in children: relation to clinical and radiological findings. Pediatr Nephrol, 6: 328, 1992 9. Stokland, E., Hellstrom, M., Jacobsson, B., Jodal, U., Lundgren, P. and Sixt, R.: Early 99mTc dimercaptosuccinic acid (DMSA) scintigraphy in symptomatic first-time urinary tract infection. Acta Paediatr, 85: 430, 1996 10. Jakobsson, B. and Svensson, L.: Transient pyelonephritic changes on 99mTechnetium-dimercaptosuccinic acid scan for at least five months after infection. Acta Paediatr, 86: 803, 1997 11. Goldraich, N. P. and Goldraich, I. H.: Update on dimercaptosuccinic acid renal scanning in children with urinary tract infection. Pediatr Nephrol, 9: 221, 1995 12. Patel, K., Charron, M., Hoberman, A., Brown, M. L. and Rogers, K. D.: Intra- and interobserver variability in interpretation of DMSA scans using a set of standardized criteria. Pediatr Radiol, 23: 506, 1993 13. Lebowitz, R. L., Olbing, H., Parkkulainen, K. V., Smellie, J. M. and Tamminen-Mobius, T. E.: International system of radiographic grading of vesicoureteric reflux. International Reflux Study in Children. Pediatr Radiol, 15: 105, 1985 14. Smellie, J. M., Edwards, D., Normand, I. C. and Prescod, N.: Effect of vesicoureteric reflux on renal growth in children with urinary tract infection. Arch Dis Child, 56: 593, 1981 15. Jodal, U. and Lindberg, U.: Guidelines for management of children with urinary tract infection and vesicoureteric reflux. Recommendations from a Swedish state-of-the-art conference. Swedish Medical Research Council. Acta Paediatr, suppl., 88: 87, 1999 16. Majd, M., Rushton, H. G., Jantausch, B. and Wiedermann, B. L.: Relationship among vesicoureteral reflux, P-fimbriated Escherichia coli, and acute pyelonephritis in children with febrile urinary tract infection. J Pediatr, 119: 578, 1991 17. Risdon, R. A., Godley, M. L., Parkhouse, H. F., Gordon, I. and Ransley, P. G.: Renal pathology and the 99mTc-DMSA image during the evolution of the early pyelonephritic scar: an experimental study. J Urol, 151: 767, 1994 18. Risdon, R. A.: The small scarred kidney in childhood. Pediatr Nephrol, 7: 361, 1993 19. Stark, H.: Urinary tract infections in girls: the cost-effectiveness of currently recommended investigative routines. Pediatr Nephrol, 11: 174, 1997 20. 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 EDITORIAL COMMENT The authors performed a prospective cohort analysis of children who presented with clinical signs and symptoms of acute pyelonephritis (APN), with APN confirmed by a positive urine culture and DMSA renal scintigraphy. Renal scans were reviewed by 2 radiologists who were blinded to each other’s interpretation as well as the clinical scenario. Cystography was performed in each child to diagnose VUR, and DMSA renal scans were repeated at 3 months to determine which acute lesions had resolved or formed scars. These acute lesions and scars were measured and their sizes expressed as a percentage of individual kidney total surface area. The means of these values for the various groups studied were then compared, with stratification being made on the basis of age, presence or absence of reflux, reflux grade, history of urinary tract infection and the presence of voiding dysfunction. The observations of the authors provide additional support for the growing use of DMSA renal scans as the initial study performed when evaluating patients following APN, thus, challenging the clas-
IMPACT OF VESICOURETERAL REFLUX ON RENAL LESIONS AFTER PYELONEPHRITIS sic diagnostic algorithm, in which renal sonography and voiding cystography are the initial studies performed to evaluate all cases of APN. This approach first identifies those at risk for acquired renal scarring, with only at risk individuals undergoing further diagnostic evaluation.1 As in previous reports, the majority of patients presenting with acute inflammatory lesions did not have VUR.2 Of the 28% of total renal units that were associated with VUR 60% were low grade, with the remainder being grades III and IV. The frequency of renal scarring increased with the severity of reflux. However, even in those patients without VUR the incidence of new renal scars was 50%, compared with 78% in those with grades III and IV VUR. Thus, if DMSA scans were done initially, half of the children without reflux would have been spared undergoing cystography, since they did not have renal scars. However, not all cases of VUR would be detected by this approach, since acute lesions resolved completely in 22% of those with grades III and IV VUR. These data are almost identical to those reported by Hansson et al, who would also have missed 20% of grade III to V VUR.1 One obvious question is whether any threshold size exists, for either acute lesions or renal scars, below which cystography is unnecessary. This information could improve the sensitivity or specificity of this approach for identifying children likely to have VUR. The authors address this issue by correlating the size of acute lesions and renal scars with the presence of VUR and VUR grade. Based on statistical analysis of the means for each group, the authors report that a quantitatively greater involvement of renal parenchyma occurs acutely in the presence of VUR than in its absence. However, their data reveal that severe acute lesions may be identified in a significant proportion of children without reflux. Thus, acute lesions
575
may not provide additional information to stratify accurately patients into those likely to have VUR and those not. However, renal scarring appears to be a better discriminator, since large lesions (defined as greater than 25% of total surface area) were much more likely to be associated with VUR, particularly grades III and IV. Again, use of a strict size criterion would exclude from cystography some children with grades III and IV VUR who are fortunate enough to have complete resolution of the acute parenchyma inflammatory lesion. This study is important, since it underscores the importance of infection as the primary determinant of renal scarring, with reflux perhaps being one of many possible additional risk factors. As future prospective studies on the pathogenesis of renal scarring are performed, it is likely that a much more tailored evaluation and management scheme will emerge that will decrease to a minimum unnecessary studies, unnecessary antibiotic use and unnecessary surgical procedures. Hans Pohl Division of Urology Children’s National Medical Center Washington, D.C. 1. Hansson, S., Dhamey, M., Sigstrom, O., Sixt, R., Stokland, E., Wennerstrom, M. et al: Dimercapto-succinic acid scintigraphy instead of voiding cystourethrography for infants with urinary tract infection. J Urol, 172: 1071, 2004 2. 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
Vol. 173, 571–575, February 2005 Printed in U.S.A.
DOI: 10.1097/01.ju.0000151263.36909.91
IMPACT OF VESICOURETERAL REFLUX ON THE SIZE OF RENAL LESIONS AFTER AN EPISODE OF ACUTE PYELONEPHRITIS ´ LEZ,* JEAN-PIERRE PAPAZYAN ELSA GONZA
AND
ERIC GIRARDIN
From the Paediatric Nephrology Unit, Department of Paediatrics, Children’s Hospital (EG, EG) and Division of Nuclear Medicine, Department of Radiology, Geneva University Hospital (J-PP), Geneva, Switzerland
ABSTRACT
Purpose: We determined the impact of vesicoureteral reflux (VUR) on the size of renal lesions in children after an episode of acute pyelonephritis. Materials and Methods: A total of 161 children (mean age 2.44 years) with acute pyelonephritis were studied. All had renal lesions on dimercapto-succinic acid scintigraphy done at admission to the hospital. A second dimercapto-succinic acid scan was performed at 3 months. Voiding cystourethrography was done at 6 weeks and VUR was graded I to V. For each renal unit layouts of renal lesions were drawn, and the damage surface was calculated and reported for the total surface of the kidney. Results: Mean size of acute lesions and scars increased with severity of reflux (p ⬍0.0001), with an important overlap of individual values. Mean size of renal scars in the group of renal units with acute lesions was 5.8% ⫾ 8.5% in patients without VUR, 9.9% ⫾ 7.3% in those with grade I reflux, 7.7% ⫾ 11.0% in those with grade II reflux, 17.7% ⫾ 14.7% in those with grade III reflux and 17.4% ⫾ 27.7% in those with grade IV reflux (p ⬍0.001). The size of renal lesions decreased significantly with time. The rate of regression of lesions decreased with increasing reflux. When analyzed according to 3 age groups sizes of scars increased significantly with age. Conclusions: VUR has an impact on the size of renal lesions after an episode of pyelonephritis. Children with a grade III or IV reflux are more likely to have larger renal scars. On the other hand, acute lesions of important size may develop even in the absence of VUR. KEY WORDS: pyelonephritis, cicatrix, vesico-ureteral reflux, technetium Tc 99m dimercaptosuccinic acid
Acute pyelonephritis in children often leads to irreversible lesions of renal parenchyma that are still present several months after the acute episode and defined as renal scars.1 It has been shown that the inflammatory process, which appears early in the course of the disease, is responsible for scar formation.2 Children with renal lesions after repeated episodes of acute pyelonephritis are at risk for development of hypertension and impairment of renal function.3–5 In pediatric pyelonephritis acute lesions of the kidney on dimercapto-succinic acid (DMSA) scan are observed in 42% to 92% of patients, leading to permanent renal damage in 36% to 68%.6 –10 The gold standard for the diagnosis of these lesions is DMSA scintigraphy, which has good sensitivity and specificity.6, 11 Vesicoureteral reflux (VUR) is classically considered a risk factor for the development of renal scarring but the development of scars has been observed to occur in the absence of VUR.6 –9 This finding raises the question of the exact correlation between VUR and renal lesions, and of the importance of VUR in the etiology of renal scarring. In the literature there is a lack of information concerning the size of renal lesions on DMSA scan. Several studies demonstrate DMSA results as positive or negative. A positive DMSA scan may correspond to small lesions of limited clinical significance. Thus, it is important to measure the size of renal lesions on DMSA scan, especially if one wants to correlate renal damage with potential risk factors such as VUR. The aim of this study was to determine the impact of VUR on the size of acute renal lesions and resultant renal scars. For this purpose the frequency of renal lesions and their
precise size on DMSA scintigraphy were analyzed during acute episodes and at 3 months. METHODS
Patients with clinically suspected pyelonephritis had a urine sample taken for culture and a blood sample for blood count, C-reactive protein determination and culture. Clinical suspicion was defined as fever greater than 38.5C, general signs of feeding difficulties, vomiting, irritability and abdominal or flank pain in older children. If the urine culture was positive, renal ultrasound and DMSA renal scan were performed in the acute phase (3 to 5 days). All patients presented initially between 1994 and 1997. Patients were included in the study if they had a positive urine culture and an acute lesion on the first DMSA renal scintigraphy. A positive urine culture was defined as 100,000 colonies per ml or greater with 1 type of bacteria (2 if the bacteria were Escherichia coli and Enterococcus faecalis) in a urine sample collected in a sterile bag, or any colonies if the sample was collected by suprapubic puncture of the bladder. Patients were excluded if they had obstructive renal disease, incomplete radiological files or a scintigram of less than 1 ⫻ 0.8 inches, which does not allow precise measurement of the surface of the lesions. Renal scintigraphy was performed according to standard techniques—3 to 4 hours after an intravenous injection of DMSA labeled 3.7 MBq 99mtechnetium per kilogram (with a minimum dose of 18.5 MBq and a maximum dose of 185 MBq) image acquisition was performed with a GCA9300A/HG gamma camera connected to a computer (Toshiba Medical Systems Co., Japan). Six different views were obtained (1 posterior, 2 posterior oblique, 1 anterior and 2 anterior oblique projections). Acute lesions were defined as a
Submitted for publication April 23, 2004. * Correspondence: Department of Paediatrics, Children’s Hospital, 6 rue Willy-Donze´, 1211 Gene`ve 14, Switzerland (e-mail: gonzaea2@ hotmail.com). 571
572
IMPACT OF VESICOURETERAL REFLUX ON RENAL LESIONS AFTER PYELONEPHRITIS
decreased uptake of isotope in at least 2 different views. Scars were defined as persistent changes in the same location as in the first DMSA scan. Every image was independently assessed by 2 experienced radiologists who used standard criteria previously defined by Patel et al.12 To calculate the precise size of renal lesions, each image was assessed regardless of clinical history and presence or absence of vesicoureteral reflux. Each outline of lesion and kidney was drawn on the view of the scintigram where the lesion was the largest. If multiple lesions were present, surface areas were added together. The corresponding surfaces were calculated and compared to the total surface of the kidney. Sizes of lesions were expressed as percentage of total kidney surface. To diagnose VUR, voiding cystourethrography (VCUG) was done 6 weeks after the diagnosis. Reflux was classified according to the gradation of the International Reflux Study committee, grade I to V.13 The surfaces of renal lesions were correlated with the result of the VCUG. Size of renal lesions was analyzed on first and second DMSA scan. To study the relation between clinical characteristics, grade of reflux and renal lesions, in case of bilateral reflux and/or lesion the side where the reflux and/or lesion was greatest was considered. Acute renal lesions and scars were further analyzed according to 4 groups of patients—those with left dominant reflux (defined as left unilateral reflux or bilateral reflux with a more important grade on the left side), those with right dominant reflux (defined as right unilateral reflux or bilateral reflux with a more important grade on the right side), those with bilateral reflux of the same grade and those without reflux. Finally, renal units with acute lesions were analyzed according to 3 age groups—less than 1 year, 1 to 5 years and more than 5 years. Results were given in terms of renal units as mean ⫾ standard error. Differences between percentages were analyzed using the chi-square test. Renal lesion sizes were compared using the Kruskal-Wallis nonparametric test and Dunn’s posttest. Differences between means were analyzed using the Mann-Whitney U test and Wilcoxon rank sum test for paired data. The effect of reflux on the progression of lesions and the effect of clinical characteristics on the size of scars were analyzed using analysis of variance. Statistical analysis was performed using the Stata statistical software package (StataCorp LP, College Station, Texas) and Prism software (GraphPad Software, Inc., San Diego, California). RESULTS
Characteristics of study population. A total of 161 children (age range birth to 11.2 years, 120 girls and 41 boys) were included in this study. Of the boys 78% were younger than 1 year at the time of diagnosis. Mean age was 1.17 years (median 0.47) for boys and 2.87 years (1.82) for girls. Clinical data are outlined in table 1. The impact of the clinical parameters on the size of renal scars was studied using analysis of variance. The presence of a previous urinary tract infection (UTI, 12 patients) and abnormal voiding habits (9) were found to be significantly related to the size of renal scars (p ⬍0.001 and p ⫽ 0.04, respectively). Mean size of renal scars in the group of patients with previous UTI was 19.1% ⫾ 20.6% compared to 8.1% ⫾ 10.5% in the other patients (p ⫽ 0.038).
TABLE 1. Characteristics of study population
Frequency of renal lesions and VUR. According to the inclusion criteria, each child had a renal lesion on the first DMSA scan. In terms of renal units, 197 of 322 renal units (61%) had lesions on the first DMSA scan. On the second DMSA scan performed at 3 months 80% of renal lesions exhibited partial or total regression, and a renal scar was present in 104 of 322 renal units (32%). Of 322 renal units vesicoureteral reflux was diagnosed in 89 (28%) on VCUG at 6 weeks. Grades of reflux were distributed as follows—15 renal units (17%) grade I, 38 (43%) grade II, 26 (29%) grade III and 10 (11%) grade IV. Among renal units with acute lesions 47% in the group without reflux had renal scars, 60% had scars in the group with reflux grades I and II, and 78% had scars in the group with reflux grades III and IV. The frequency of renal scars increased with the severity of reflux (p ⫽ 0.003). Size of renal lesions and VUR. When analyzed according to the grade of reflux an overlap of individual values was observed. Nevertheless, the mean size of acute renal lesions increased when reflux was present (Mann-Whitney U test between renal units with and without reflux, p ⬍0.0001, fig. 1, A). Of the subgroup of renal units with acute lesions the mean size of renal lesions was 19.1% ⫾ 12.1% in the group without VUR, 25.2% ⫾ 8.8% in the group with grade I VUR, 18.0% ⫾ 10.4% in the group with grade II VUR, 24.0% ⫾ 12.4% in the group with grade III VUR and 27.6% ⫾ 25.4% in the group with grade IV VUR. Overlap was also observed for renal scars. Mean size increased with severity of reflux (Kruskal-Wallis nonparametric test, p ⬍0.0001, fig. 1, B). Of the subgroup of renal units with acute lesions the mean size of renal scars was 5.8% ⫾ 8.5% in the group without VUR, 9.9% ⫾ 7.3% in the group with grade I VUR, 7.7% ⫾ 11.0% in the group with grade II VUR, 17.7% ⫾ 14.7% in the group with grade III VUR and 17.4% ⫾ 27.7% in the group with grade IV VUR (KruskalWallis nonparametric test, p ⬍0.001). The risk for a child having a renal scar of more than 25% of the surface of the kidney increased with grade of reflux (p ⫽ 0.015, table 2). When analyzing regression of lesions at 3 months in the group of renal units with an acute lesion the size of renal lesions decreased significantly with time in all groups. The grade of reflux affected significantly the rate of regression of lesions. The greater the reflux, the lesser the regression (ANOVA, p ⬍0.001, table 3). To highlight the correlation between VUR and renal lesions for a given patient, renal lesions were analyzed by subtracting the size of lesions of 1 side from the other (right minus left) for each patient and analyzed according to 4 groups—patients with left dominant reflux, right dominant reflux, reflux of same grade and without reflux. In patients with left dominant reflux the mean size of acute renal lesions and scars on the left side was significantly larger than in those with right dominant reflux (p ⬍0.05 and p ⬍0.01, respectively), indicating that renal lesions were more extensive when reflux was present on the same side. Size of renal lesions and age. The frequency and size of renal scars were analyzed according to 3 age groups— younger than 1 year, 1 to 5 years and older than 5 years. The frequency of renal scarring increased with age (p ⫽ 0.0003). When considering only renal units with an acute lesion sizes of scars increased significantly with age (p ⬍0.0001), and posttest revealed statistically significant differences between the group of patients younger than 1 year and the other 2 age groups (p ⬍0.001, fig. 2).
No. Pts Abnormal voiding habits 9 Abnormal bowel habits 7 Previous UTI 12 Abnormal urethra on VCUG 6 Abnormal bladder on VCUG* 5 * One patient had bladder thickening and 4 had bladder diverticuli.
DISCUSSION
Renal lesions may lead to hypertension and renal function impairment. Thus, one of the main goals of pyelonephritis management is to decrease the occurrence of renal scars. As reflux has traditionally been considered a risk factor for renal
IMPACT OF VESICOURETERAL REFLUX ON RENAL LESIONS AFTER PYELONEPHRITIS
573
FIG. 2. Size of renal scars in group of renal units with acute lesions according to age—younger than 1 year, 1 to 5 years and greater than 5 years (p ⬍0.001).
FIG. 1. Size of renal lesions on DMSA scintigraphy according to grade of reflux. A, acute lesions are demonstrated on DMSA scan. B, renal scars are shown on DMSA scan performed at 3 months. TABLE 2. Risk of scar more than 25% of total surface of kidney according to reflux grade No. pts without reflux/total No. (%) No. pts with grades I ⫹ II reflux (%) No. pts with grades III ⫹ IV reflux (%)
3/50 (6) 3/22 (14) 8/20 (40)
TABLE 3. Regression of renal lesions related to reflux grade % Acute Lesions
% Scars
p Value
No reflux 19.1 ⫾ 12.1 5.7 ⫾ 8.4 ⬍0.0001 Grades I and II reflux 20.1 ⫾ 10.4 8.3 ⫾ 10.1 ⬍0.0001 Grades III and IV reflux 24.0 ⫾ 16.2 18.0 ⫾ 17.6 0.0051 Analysis of variance showed that time (p ⬍0.001) and degree of reflux (p ⬍0.001) influenced size of lesions.
lesions,14, 15 considerable efforts were undertaken to diagnose and correct VUR. Nevertheless, several studies indicate that scars are fre-
quently observed in the absence of reflux.6, 8, 16 This observation leads to the question of the exact correlation between reflux and renal lesions. The answer has a direct impact on the clinical management of acute pyelonephritis and on radiological investigations to be performed after an episode of pyelonephritis. DMSA scintigraphy is considered the gold standard for detecting renal lesions. Several studies show DMSA results as positive or negative. Thus, a positive DMSA scan may correspond to small lesions of limited clinical significance. To predict the outcome of acute lesions, several authors have classified the abnormalities demonstrated on the DMSA scan. Goldraich and Goldraich proposed a 4-grade classification depending on the number of lesions, the persistence of normal parenchyma between lesions and the atrophic aspect of the kidney.11 Risdon et al observed that with a classification of 3 types the majority of type C lesions, corresponding to focal defects with complete loss of uptake and loss of renal outline, persisted as irreversible scarring.17 Jakobsson et al quantified lesions on DMSA scan and found that reflux of grade III or more was associated with significantly greater defects (defined as more than 10% of the kidney with DMSA uptake less than 2 SD of the control kidney).8 In the literature there is a lack of information concerning the exact size of renal lesions on DMSA scan. In the present study renal lesions were measured in terms of percentage of the area of the kidney and correlated with the grade of reflux determined by DMSA scintigraphy. Each child enrolled in this study had a renal lesion on DMSA scan done at the time of diagnosis, and 56% had renal scars on the second scintigraphy performed at 3 months. In terms of renal units acute lesions were present in 61% initially and persisted in 53% at 3 months. At the time of the design of the study 3-month followup was considered appropriate to diagnose scars. It was subsequently revealed that some lesions may further regress by 5 months.10 Nevertheless, the frequencies observed in our study are in the range of results previously reported.6, 8, 10 Are the renal lesions observed a consequence of infection or a consequence of preexisting congenital damage? It is difficult to evaluate the origin of lesions on the basis of DMSA scintigraphy alone.18 Clues exist to indicate that congenital
574
IMPACT OF VESICOURETERAL REFLUX ON RENAL LESIONS AFTER PYELONEPHRITIS
lesions did not represent a major cause of renal lesions observed in our study population. Patients with an obstructive disease were excluded, and none of the patients included in the study had hydronephrosis or dysplasic kidneys on fetal ultrasound. Moreover, regression of lesions was observed in 80% of the cases, indicating that at least some of the lesions were related to acute inflammation. Our data confirmed the role of VUR as a risk factor for renal lesions. Mean size of renal scars increased significantly with grade of reflux, and children with grade III to IV reflux were more likely to have development of a larger scar than those with grades I and II. When data were analyzed for each individual patient sizes of renal lesions were larger when ipsilateral reflux was present. These results might underestimate the importance of VUR, as data were collected after a first episode of acute pyelonephritis for the majority of patients, and, thus, do not give information on the long-term impact of VUR on renal lesions. In our population voiding habits and a history of UTI were additional risk factors for larger scars. Longitudinal studies are needed to evaluate this impact. It has been suggested that DMSA scintigraphy might be performed only in patients in whom VCUG shows VUR.19 Our data demonstrate that when individual renal lesions were correlated with grade of reflux value overlaps did not allow prediction of size of acute renal lesions or renal scars according to the absence or presence of reflux and to its grade, nor prediction of the degree of reflux from the size of the renal lesions (fig. 1). Moreover, large acute renal lesions and scars also occurred in the absence of VUR. To evaluate the effect of age on the risk of developing scars, we analyzed our data according to 3 age groups—younger than 1 year, 1 to 5 years and older than 5 years. Previous studies have revealed controversial data concerning renal scarring and age. It is generally admitted that infants are at higher risk for renal scarring than older children but in our study the mean size of renal lesions increased with age. These results confirmed data from recent studies showing that older children are at higher risk for persistent renal lesions.10, 20 Moreover, the relationship between reflux and size of renal scars was similar when considering only children older than 1 year (data not shown). CONCLUSIONS
These data demonstrate that VUR has an impact on the size of renal lesions after an episode of pyelonephritis, and that children with grade III or IV reflux are more likely to have development of larger renal scars. On the other hand, acute lesions of important size may develop even in the absence of VUR. In the treatment of acute pyelonephritis in children, the choice of the radiological investigation to be performed, DMSA scintigraphy and/or VCUG, should be made knowing that one investigation cannot predict the result of the other. Bernadette Mermillod provided advice on statistical analysis. REFERENCES
1. Jakobsson, B., Berg, U. and Svensson, L.: Renal scarring after acute pyelonephritis. Arch Dis Child, 70: 111, 1994 2. Bille, J. and Glauser, M. P.: Protection against chronic pyelonephritis in rats by suppression of acute suppuration: effect of colchicine and neutropenia. J Infect Dis, 146: 220, 1982 3. Jacobson, S. H., Eklof, O., Eriksson, C. G., Lins, L. E., Tidgren, B. and Winberg, J.: Development of hypertension and uraemia after pyelonephritis in childhood: 27 year follow up. BMJ, 299: 703, 1989 4. Smellie, J. M., Prescod, N. P., Shaw, P. J., Risdon, R. A., Bryant, T. N.: Childhood reflux and urinary infection: a follow-up of 10 – 41 years in 226 adults. Pediatr Nephrol, 12: 727, 1998
5. Patzer, L., Seeman, T., Luck, C., Wuhl, E., Janda, J. and Misselwitz, J.: Day- and night-time blood pressure elevation in children with higher grades of renal scarring. J Pediatr, 142: 117, 2003 6. Majd, M. and Rushton, H. G.: Renal cortical scintigraphy in the diagnosis of acute pyelonephritis. Semin Nucl Med, 22: 98, 1992 7. Benador, D., Benador, N., Slosman, D. O., Nussle, D., Mermillod, B. and Girardin, E.: Cortical scintigraphy in the evaluation of renal parenchymal changes in children with pyelonephritis. J Pediatr, 124: 17, 1994 8. Jakobsson, B., Nolstedt, L., Svensson, L., Soderlundh, S. and Berg, U.: 99mTechnetium-dimercaptosuccinic acid scan in the diagnosis of acute pyelonephritis in children: relation to clinical and radiological findings. Pediatr Nephrol, 6: 328, 1992 9. Stokland, E., Hellstrom, M., Jacobsson, B., Jodal, U., Lundgren, P. and Sixt, R.: Early 99mTc dimercaptosuccinic acid (DMSA) scintigraphy in symptomatic first-time urinary tract infection. Acta Paediatr, 85: 430, 1996 10. Jakobsson, B. and Svensson, L.: Transient pyelonephritic changes on 99mTechnetium-dimercaptosuccinic acid scan for at least five months after infection. Acta Paediatr, 86: 803, 1997 11. Goldraich, N. P. and Goldraich, I. H.: Update on dimercaptosuccinic acid renal scanning in children with urinary tract infection. Pediatr Nephrol, 9: 221, 1995 12. Patel, K., Charron, M., Hoberman, A., Brown, M. L. and Rogers, K. D.: Intra- and interobserver variability in interpretation of DMSA scans using a set of standardized criteria. Pediatr Radiol, 23: 506, 1993 13. Lebowitz, R. L., Olbing, H., Parkkulainen, K. V., Smellie, J. M. and Tamminen-Mobius, T. E.: International system of radiographic grading of vesicoureteric reflux. International Reflux Study in Children. Pediatr Radiol, 15: 105, 1985 14. Smellie, J. M., Edwards, D., Normand, I. C. and Prescod, N.: Effect of vesicoureteric reflux on renal growth in children with urinary tract infection. Arch Dis Child, 56: 593, 1981 15. Jodal, U. and Lindberg, U.: Guidelines for management of children with urinary tract infection and vesicoureteric reflux. Recommendations from a Swedish state-of-the-art conference. Swedish Medical Research Council. Acta Paediatr, suppl., 88: 87, 1999 16. Majd, M., Rushton, H. G., Jantausch, B. and Wiedermann, B. L.: Relationship among vesicoureteral reflux, P-fimbriated Escherichia coli, and acute pyelonephritis in children with febrile urinary tract infection. J Pediatr, 119: 578, 1991 17. Risdon, R. A., Godley, M. L., Parkhouse, H. F., Gordon, I. and Ransley, P. G.: Renal pathology and the 99mTc-DMSA image during the evolution of the early pyelonephritic scar: an experimental study. J Urol, 151: 767, 1994 18. Risdon, R. A.: The small scarred kidney in childhood. Pediatr Nephrol, 7: 361, 1993 19. Stark, H.: Urinary tract infections in girls: the cost-effectiveness of currently recommended investigative routines. Pediatr Nephrol, 11: 174, 1997 20. 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 EDITORIAL COMMENT The authors performed a prospective cohort analysis of children who presented with clinical signs and symptoms of acute pyelonephritis (APN), with APN confirmed by a positive urine culture and DMSA renal scintigraphy. Renal scans were reviewed by 2 radiologists who were blinded to each other’s interpretation as well as the clinical scenario. Cystography was performed in each child to diagnose VUR, and DMSA renal scans were repeated at 3 months to determine which acute lesions had resolved or formed scars. These acute lesions and scars were measured and their sizes expressed as a percentage of individual kidney total surface area. The means of these values for the various groups studied were then compared, with stratification being made on the basis of age, presence or absence of reflux, reflux grade, history of urinary tract infection and the presence of voiding dysfunction. The observations of the authors provide additional support for the growing use of DMSA renal scans as the initial study performed when evaluating patients following APN, thus, challenging the clas-
IMPACT OF VESICOURETERAL REFLUX ON RENAL LESIONS AFTER PYELONEPHRITIS sic diagnostic algorithm, in which renal sonography and voiding cystography are the initial studies performed to evaluate all cases of APN. This approach first identifies those at risk for acquired renal scarring, with only at risk individuals undergoing further diagnostic evaluation.1 As in previous reports, the majority of patients presenting with acute inflammatory lesions did not have VUR.2 Of the 28% of total renal units that were associated with VUR 60% were low grade, with the remainder being grades III and IV. The frequency of renal scarring increased with the severity of reflux. However, even in those patients without VUR the incidence of new renal scars was 50%, compared with 78% in those with grades III and IV VUR. Thus, if DMSA scans were done initially, half of the children without reflux would have been spared undergoing cystography, since they did not have renal scars. However, not all cases of VUR would be detected by this approach, since acute lesions resolved completely in 22% of those with grades III and IV VUR. These data are almost identical to those reported by Hansson et al, who would also have missed 20% of grade III to V VUR.1 One obvious question is whether any threshold size exists, for either acute lesions or renal scars, below which cystography is unnecessary. This information could improve the sensitivity or specificity of this approach for identifying children likely to have VUR. The authors address this issue by correlating the size of acute lesions and renal scars with the presence of VUR and VUR grade. Based on statistical analysis of the means for each group, the authors report that a quantitatively greater involvement of renal parenchyma occurs acutely in the presence of VUR than in its absence. However, their data reveal that severe acute lesions may be identified in a significant proportion of children without reflux. Thus, acute lesions
575
may not provide additional information to stratify accurately patients into those likely to have VUR and those not. However, renal scarring appears to be a better discriminator, since large lesions (defined as greater than 25% of total surface area) were much more likely to be associated with VUR, particularly grades III and IV. Again, use of a strict size criterion would exclude from cystography some children with grades III and IV VUR who are fortunate enough to have complete resolution of the acute parenchyma inflammatory lesion. This study is important, since it underscores the importance of infection as the primary determinant of renal scarring, with reflux perhaps being one of many possible additional risk factors. As future prospective studies on the pathogenesis of renal scarring are performed, it is likely that a much more tailored evaluation and management scheme will emerge that will decrease to a minimum unnecessary studies, unnecessary antibiotic use and unnecessary surgical procedures. Hans Pohl Division of Urology Children’s National Medical Center Washington, D.C. 1. Hansson, S., Dhamey, M., Sigstrom, O., Sixt, R., Stokland, E., Wennerstrom, M. et al: Dimercapto-succinic acid scintigraphy instead of voiding cystourethrography for infants with urinary tract infection. J Urol, 172: 1071, 2004 2. 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