 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
COMMENTARY |
Children's Hospital of Pittsburgh, Pittsburgh, Pennsylvania
Abbreviations: UTI, urinary tract infection • VCUG, voiding cystourethrogram • APN, acute pyelonephritis • DMSA, dimercaptosuccinic acid • VUR, vesicoureteral reflux
Urinary tract infection (UTI) is now the most common serious bacterial infection that occurs in infancy and early childhood in the United States.1 The prominence of UTI is a result, in great part, of the dramatic reduction of the previously common manifestations of infection caused by Haemophilus influenzae type b and Streptococcus pneumoniae due to the introduction and near-universal dissemination of effective conjugate vaccines in childhood.
The current standard of care in the management of UTI (especially for young infants and children) includes the performance of imaging procedures.2 Although some controversy has surrounded the selection of imaging procedures, the potential menu of choices includes renal ultrasound, voiding cystourethrogram (VCUG), and renal scintigram. Formerly, the intravenous pyelogram was included among the choices but has been replaced, for most intents and purposes, by renal ultrasound.
Renal ultrasound is used to assess the gross anatomy of the urinary tract and thereby detect obstructive uropathies. It is a noninvasive test that adequately describes renal size, detects dilatation and duplication of the collecting system, and gross anatomic abnormalities such as a horseshoe kidney.2 Although ultrasounds have become standard in the evaluation of children with UTI, it is worth rethinking their contribution to management. In an era when ultrasounds are performed routinely during pregnancy (and sometimes on multiple occasions), they may be superfluous in the evaluation of a child with UTI.3 Most children with congenital obstruction of the urinary tract are diagnosed in utero. If an ultrasound was performed in the latter part of gestation (beyond 30–32 weeks of gestation) at an experienced institution and found to be normal, repeating it when a young child presents with UTI has a negligible impact on management. In a recent study that evaluated intravenous versus oral treatment of 306 children between the ages of 1 and 24 months who were experiencing their first UTI, renal ultrasound findings were normal in 88% of the children. In the remaining 12%, findings were of modest significance and did not lead to any changes in management.4
Renal scintigraphy, introduced >20 years ago, may be used to confirm the diagnosis of suspected acute pyelonephritis (APN) and define the extent of parenchymal damage. Dimercaptosuccinic acid (DMSA) is an amino acid that is taken up by the renal tubular cells. When this substance or glucoheptonate is labeled with technetium, it localizes at the proximal tubules after intravenous injection. If imaging is performed 3 to 6 hours later, it will show vascular flow to the kidney and reflect tubular function. In patients with APN it is reported to have a sensitivity of nearly 90% to demonstrate an abnormality consistent with APN.5 False-negative studies may be attributed to early infection in which inflammation and dysfunction are confined to the medullary portion of the kidney, which may not be reflected by the scan using DMSA, a primarily cortical imaging agent.6 A DMSA scan can also be used to assess renal scarring when it is performed at least 6 months after the infection. Although the renal scintigram can be used to document the presence of APN and renal scarring, it is unnecessary in the management of most acute UTIs and is not recommended as part of routine evaluation.2
The contrast VCUG, to detect vesicoureteral reflux (VUR), has been considered by most, but not all, experts to be a very important study to perform. This is based on the understanding that one of the principle causes of renal scarring is reflux nephropathy. Renal damage occurs when infected urine refluxes into the kidney. Scarring can be avoided if the urine remains uninfected. VUR represents a failure of the function of the ureterovesical junction, which frequently resolves spontaneously by the age of 4 or 5 years. The rationale for performing the VCUG is to diagnose the presence of VUR and provide antimicrobial prophylaxis until the VUR either resolves spontaneously or, in cases of high degrees of VUR, is repaired surgically.7
VUR is graded on the basis of severity and designated I to V in International Study Classification (International Reflux Study Committee, 1981).2 Grade I is VUR into the distal ureter, grade II is VUR into the proximal ureter (without dilatation), and grades III, IV, and V represent VUR into the kidney with mild, moderate, and severe dilatation, respectively, of the ureters and pelvis. In general, prophylaxis is recommended for any degree of VUR. To routinely reassess the persistence of VUR, it is recommended that a radionuclide study of reflux be done at regular intervals of every 6 to 12 months. A radionuclide VCUG rather than a contrast VCUG is preferred, because the radiation exposure is substantially reduced with the radionuclide study and its sensitivity is adequate to determine the presence or absence of VUR.2
When a VCUG demonstrates VUR, antimicrobial prophylaxis is prescribed on the premise that it is beneficial in the prevention of recurrent UTI and scarring, which may lead to end-stage renal disease. However, almost all previously reported studies performed on children with VUR have compared antibiotic prophylaxis to the combination of antibiotic prophylaxis and surgery or have compared antibiotic prophylaxis to surgery alone.8,9 Only one small investigation (cited in a systematic review of randomized, controlled trials of the effect of various interventions in patients with VUR), which showed no difference in the risk for UTI or renal damage, compared intermittent prophylaxis with continuous prophylaxis and no treatment.8 There have been no large, randomized, prospective trial reports that have compared the use of continuous urinary antibiotic prophylaxis to observation and prompt treatment of acute recurrent episodes of UTI as they occur.
The aims of the study reported by Garin et al10 (in this issue of Pediatrics) were to (1) evaluate the role of VUR in increasing the frequency and severity of UTI and renal parenchymal damage in patients after an episode of APN and (2) determine if antibiotic prophylaxis reduces the frequency and/or severity of UTI and/or prevents renal parenchymal damage in patients with mild to moderate degrees of VUR.
Patients with APN who were between 3 months and 18 years of age were studied. The diagnosis of APN was substantiated by the finding of significant pyuria (>10 white blood cells per high-power field, spun sediment) and bacteriuria (>105 colony-forming units/mL) in febrile children with typical findings of APN on DMSA scintigraphy. A VCUG was performed shortly after diagnosis, although the precise timing of this study was not indicated. After treatment of patients with an antibiotic selected at the discretion of their physician, children with grade I, II, or III VUR were randomly assigned to either receive or not receive antimicrobial prophylaxis with either sulfamethoxazole/trimethoprim or nitrofurantoin. A placebo was not administered to the control group. Standardizing the choice of antibiotic treatment and antimicrobial prophylaxis would have strengthened the study design, as would the inclusion of a placebo for children who did not receive prophylaxis. Primary end points were rates and types of recurrence of UTI and development of renal scars. Unfortunately, the primary analysis was performed only on patients who completed the 1-year follow-up; an intention-to-treat analysis was not performed.
Two hundred thirty-six patients were enrolled. Low-grade VUR (grades I–III) was confirmed in 113 (48%) patients; no information was provided regarding the frequency of VUR of grades IV or V. The resolution of VUR during the year of study was similar to reported rates in other investigations and showed a higher likelihood of resolution when the grade of VUR was lowest and unilateral. The rate of resolution did not differ significantly in the groups with or without the use of urinary antibiotic prophylaxis.
The overall rate of recurrence of UTI including all groups was 20.1%. Although the rates of recurrence (among patients with or without VUR) were similar in those not taking prophylaxis, for patients who were taking antibiotic prophylaxis, the recurrence rate of 8.8% among patients without VUR compared with 23.6% in those with VUR approached statistical significance (P = .063). Most recurrent infections occurred between 9 and 12 months after the initial infection and seemed to be episodes of cystitis rather than APN based on reports of a normal DMSA scan performed at the time of each recurrence (it is possible that these events represent episodes of APN detected earlier than usual because of heightened awareness in parents of children with a previous UTI). The overall recurrence rate for APN was 5.5%; recurrences were observed twice as frequently in patients with VUR than in those without (P = .378). Of 8 patients with VUR who developed recurrent APN, 6 had grade III VUR. Recurrent APN was documented in 7 patients receiving urinary antibiotic prophylaxis compared with only 1 of the patients on no prophylaxis (P = .029). In all 7 cases, the offending bacteria showed resistance to the antibiotics used for prophylaxis. Of 218 patients, 13 (5.9%) developed renal scars identified by DMSA scan, including 7 with VUR (6.2%) and 6 without VUR (5.7%).
The authors concluded that 1 year after an acute episode of APN, mild to moderate VUR did not seem to increase the overall incidence of recurrent UTI, APN, or renal scarring. Furthermore, their data do not support a role for antibiotic prophylaxis to prevent recurrence of infection or the development of renal scars (in fact, in this study, prophylaxis increased the chance of developing APN!).
The first conclusion is strongly supported by other investigators.11–14 Linshaw11 points out that because UTIs are initiated when urothelial cell receptors allow bacterial attachment, there is no compelling reason to expect that VUR should predispose patients to UTI. Although there is no evidence that VUR predisposes to an increased incidence of UTI, it is conceivable that when infection does occur, it is more likely to be APN rather than cystitis. Limited data from this study, although not statistically significant, support this possibility.10 Whereas VUR is clearly a risk factor for renal scarring in the presence of UTI, there is no clear causal relationship between VUR and scarring when infection is not present.11,15 In numerous studies, scarring has been documented in children with either low degrees of or no VUR at all,15,16 whereas no scarring has been observed in children with high degrees of sterile VUR.9,15
Why wouldn't antibiotic prophylaxis prevent recurrent infection? The source of bacteria that cause UTI are derived from the gastrointestinal tract. Rectal flora colonize the periurethral area and ultimately gain access to the urethra and bladder. Frequent and complete voiding are the mechanisms that primarily are responsible for maintaining the bladder and upper urinary tracts free of infection. When these mechanisms fail, a UTI develops. Having an effective antimicrobial level in the bladder urine should theoretically prevent the development of a UTI. Again, there is a paucity of data to address this issue. A review of the literature published in 2001 disclosed only 2 small studies performed >25 years ago that addressed this issue in children without VUR.17–19
There are 2 potential barriers to the success of antimicrobial prophylaxis for UTIs. The first potential barrier is long-term adherence to the regimen (complicated by noncompliance and potential adverse effects of the drugs); the second is the emergence of antimicrobial resistance.
Only 2 antimicrobial agents are generally recommended for prophylaxis of UTI: sulfamethoxazole-trimethoprim and nitrofurantoin. It is predictable that other antimicrobial agents that are satisfactory for treatment (amoxicillin, cephalexin, cefuroxime, ciprofloxacin, etc) will not suffice for prophylaxis because of their pharmacokinetic profile (characteristics of absorption), which permits exposure of Gram-negative bacilli to low concentrations of antimicrobial agents present in the colon after oral administration. Invariably, the population of coliforms that remain a few weeks after initiating these antimicrobial agents are those that are inherently resistant or have become resistant to the prescribed drug. The absorption of nitrofurantoin and sulfamethoxazole-trimethoprim high in the gastrointestinal tract protects these antimicrobial agents, to some extent, for prolonged periods because the native flora of the colon are not exposed.19 Nonetheless, the observation that most reinfections in this series occurred at 9 to 12 months requires both long-term adherence and long-term staying power for the antimicrobial agent in question. Unfortunately, in some geographic areas sulfamethoxazole-trimethoprim is no longer an ideal agent because of the prevalence of endemic resistance among Gram-negative flora; other studies have shown the emergence of resistance after short periods of usage.20 Both nitrofurantoin and sulfamethoxazole-trimethoprim, although well-tolerated in general, may be associated with gastrointestinal adverse effects and dermatologic problems, respectively.21
Another consideration is whether prophylaxis (if it were possible to maintain activity against the likely pathogens) is substantially more effective in preventing renal scars than very early treatment of acute infections. Theoretically, effective treatment undertaken at the earliest indication of infection might either abort or minimize the inflammatory response and thereby prevent scarring (although minor morbidity associated with acute mild infections might not be avoided). This question cannot be answered until there is a prophylactic strategy that actually maintains activity against suspected pathogens.
The solution to this problem might revolve around systematic alternation of antimicrobial agents (every 2–4 weeks) for prophylaxis, as has been recommended for prevention of respiratory infections in children with common variable immunodeficiency.22 Although this was not successful in one recent study of prophylaxis for UTI, others have suggested its potential effectiveness.20,21 An attractive alternative would be to use a nonantibiotic antiinfective such as methenamine mandelate. This chemical creates a sterile bladder urine pharmacologically by converting methenamine to formic acid if the urine can be acidified to a pH of <6. Practically speaking, acidification such as this may be difficult to accomplish, but a similar strategy would be very attractive because it does not expose the host to antimicrobial agents.
Garin et al10 have performed a very important study and have provided compelling data indicating that urinary tract prophylaxis with currently available agents in children with low grades of VUR (I, II, and III) does not seem to be beneficial. This investigation will surely stimulate additional studies to continue to clarify this issue. However, we must remember that these conclusions may not apply to children with higher degrees of reflux. Until such time that there are data to address the usefulness of prophylaxis in these latter cases, we must continue to recommend the performance of the VCUG to determine the presence of VUR and search for strategies to keep the urine free of infection in children with high degrees of VUR.
 |
FOOTNOTES |
|---|
Address correspondence to Ellen R. Wald, MD, Department of Pediatrics, Box 4108, University of Wisconsin Children's Hospital, 600 Highland Ave, Madison, WI 53792. E-mail: erwald{at}wisc.edu
The author has indicated she has no financial relationships relevant to this article to disclose.
|
REFERENCES |
|---|
 TOP REFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  C.-H. Cheng, M.-H. Tsai, Y.-C. Huang, L.-H. Su, Y.-K. Tsau, C.-J. Lin, C.-H. Chiu, and T.-Y. Lin Antibiotic Resistance Patterns of Community-Acquired Urinary Tract Infections in Children With Vesicoureteral Reflux Receiving Prophylactic Antibiotic Therapy Pediatrics, December 1, 2008; 122(6): 1212 - 1217. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. B. Newman Much Pain, Little Gain From Voiding Cystourethrograms After Urinary Tract Infection Pediatrics, November 1, 2006; 118(5): 2251 - 2251. [Full Text] [PDF]
|
|
|
|
 |
|
 JOURNAL WATCH PEDIATRICS & ADOLESCENT MEDICINE Arch. Dis. Child., August 1, 2006; 91(8): 712 - 713. [Full Text] [PDF]
|
|
|
|
 |
|
 Antibiotic Prophylaxis for Vesicoureteral Reflux May Not Be Necessary Journal Watch Pediatrics and Adolescent Medicine, April 21, 2006; 2006(421): 3 - 3. [Full Text]
|
|
|
|
 |
|
 UTIs, Vesicoureteral Reflux, and Antibiotic Prophylaxis Journal Watch (General), March 28, 2006; 2006(328): 5 - 5. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2006 American Academy of Pediatrics. All rights reserved. |
||
|
||
