Antibiotic Prophylaxis for Prevention of Febrile Urinary Tract Infections in Children with Vesicoureteral Reflux: A Meta-Analysis of Randomized, Controlled Trials Comparing Dilated to Nondilated Vesicoureteral Reflux Jose de Bessa, Jr.,* Flavia Cristina de Carvalho Mrad, Evilin Feitosa Mendes, Marcia Carvalho Bessa, Victor Pereira Paschoalin, Ricardo Brianezi Tiraboschi, Zein Mohamed Sammour, Cristiano Mendes Gomes, Luis H. Braga and Murillo Bastos Netto Jose From the Universidade Estadual de Feira de Santana (JdB, EFM, MCB, VPP, RBT), Feira de Santana, Hospital e Maternidade Therezinha de Jesus da Faculdade de Cieˆncias Me´dicas e da Sau´de de Juiz de Fora (Suprema) (JMBN) and Universidade Federal de Juiz de Fora (FCdCM, JMBN), Juiz de Fora and Universidade de Sa˜o Paulo, Sa˜o Paulo, Brazil, and Division of Urology, Department of Surgery, McMaster University (LHB), Hamilton, Ontario, Canada
Abbreviations and Acronyms AAP ¼ American Academy of Pediatrics ARR ¼ absolute risk reduction CAP ¼ continuous antibiotic prophylaxis NNT ¼ number needed to treat RIVUR ¼ Randomized Intervention for Children with Vesicoureteral Reflux UTI ¼ urinary tract infection VUR ¼ vesicoureteral reflux Accepted for publication October 10, 2014. * Correspondence: Universidade Estadual de Feira de Santana, Feira de Santana, Brazil (e-mail: [email protected]).
Purpose: The followup and treatment of children with vesicoureteral reflux has been debated for many years. Antibiotic prophylaxis has a role for preventing urinary tract infection in these children. Recent studies and guidelines suggested that prophylaxis has little or no role in preventing urinary tract infection in those children, especially those with low grades (I and II) of reflux. Materials and Methods: We analyzed all published randomized, controlled trials comparing antibiotic prophylaxis vs no prophylaxis or placebo in children with vesicoureteral reflux. The children were divided into those with nondilated (grades I and II) and dilated (grades III and IV) vesicoureteral reflux. After data were analyzed the RIVUR study was published and, therefore, it was added to the analyzed data. Results: After analyzing the first published studies we found that antibiotic prophylaxis would be beneficial only in children with high grade vesicoureteral reflux. With the addition of the data in the RIVUR study these results changed. The new pooled data support antibiotic prophylaxis in all children with vesicoureteral reflux. Conclusions: Vesicoureteral reflux management is still controversial. In contrast to recently published studies and guidelines, this meta-analysis supports antibiotic prophylaxis in all children with vesicoureteral reflux regardless of reflux grade. More studies are needed to support this finding. Key Words: kidney, urinary tract infections, vesico-ureteral reflux, antibiotic prophylaxis, pyelonephritis
LOW dose, prolonged antibiotic prophylaxis has been used for more than 4 decades to prevent recurrent UTIs and consequent renal scars in infants and children with VUR.1,2 This practice was based on expert opinions
1772
j
www.jurology.com
and poorly designed, nonrandomized studies, and supported by many professional societies.3e8 More recently better designed studies questioned the real benefits of antibiotic prophylaxis in children with VUR and showed a
0022-5347/15/1935-1772/0 THE JOURNAL OF UROLOGY® © 2015 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC.
http://dx.doi.org/10.1016/j.juro.2014.10.092 Vol. 193, 1772-1777, May 2015 Printed in U.S.A.
ANTIBIOTIC PROPHYLAXIS FOR URINARY TRACT INFECTIONS IN VESICOURETERAL REFLUX
lack of convincing evidence for this practice, adding to the current controversy on this topic.9e17 These studies and others raised many questions on whether children who receive antibiotic prophylaxis are better protected from new onset pyelonephritis, which children would benefit the most from prophylaxis and whether prophylaxis really protects the kidneys from new scar formation. Therefore, other meta-analyses and systematic reviews, including a Cochrane commentary, were done with the objective of better understanding all published data on this subject.4,18,19 These studies demonstrated that many of the mentioned questions remain to be answered and strong evidence supporting this practice is still lacking. However, no previously published meta-analysis analyzed data separating high grade (dilated, grades III and IV) from low grade (nondilated, grades I and II) VUR. Recently the multicenter, randomized RIVUR trial showed that antibiotic prophylaxis was effective in decreasing recurrent UTIs in children with VUR.20 This suggested that the AAP and NICE (National Institute for Health and Care Excellence) recommendations5e8 should be reevaluated. We systematically reviewed the VUR literature to evaluate the effects of CAP to prevent recurrent UTIs in children with dilating VUR (grades III and IV) compared to those with no dilating VUR (grades I and II). We hypothesized that antibiotic prophylaxis would be effective for decreasing the number of recurrent febrile UTIs compared to no treatment in children with grades III and V VUR but not in those with grades I and II.
MATERIAL AND METHODS On November 30, 2013 we performed an electronic survey of VUR trials published in English using MEDLINEÒ, EmbaseÒ and ScopusÒ. We included all randomized and quasi-randomized, controlled trials comparing antibiotics to placebo/no treatment or comparing 2 or more antibiotics administered daily for at least 2 months to prevent recurrent UTIs in children with VUR. Studies of curative treatment, lifestyle interventions and preoperative antibiotic prophylaxis were excluded from analysis. The primary outcome was a recurrent febrile UTI. The initial review and first update were done by 3 of us (FCdCM, EFM and MCB). All selected articles were checked independently against study inclusion criteria and design by 2 groups of 3 reviewers (FCdCM, EFM and MCB or RBT, VPP and ZMS). In case of discordance other experts were consulted (JdB or JMBN). Data extraction from eligible studies was performed independently by the same reviewers with discrepancies in data extraction solved by a third referee (LHB). The internal validity of included trials was assessed by 2 reviewers using the Jadad scale.21 Our initial analysis comprised 6 articles that met study inclusion criteria. We collected data on all children in
1773
these studies who were reported to have VUR. Nondilated VUR was defined as low grade (I and II) and dilated VUR was defined as higher grade (III to IV).18 Secondary analysis was performed to compare pooled results before RIVUR publication to that of patients in RIVUR.20 A random effects model was used to estimate the risk ratio and risk difference for recurrent febrile UTIs with the 95% CI. Relative risk was calculated as the risk ratio of exposed to nonexposed, that is exposure to prophylactic antibiotics. NNT was calculated to synthesize the potential beneficial role of antibiotic prophylaxis. I2 was used to quantify heterogeneity. Meta-analysis was performed when deemed appropriate, ie when heterogeneity among studies was not considered too high. All statistical analysis was done with Comprehensive Meta-Analysis (Biostat, Englewood, New Jersey).
RESULTS Pooled data from the 6 randomized, controlled trials published until April 2014 included a total of 986 children, of whom 663 (67.2%) were female and 323 (32.8%) were male.9e14 Dilated and nondilated VUR was observed in 471 (47.7%) and 515 patients (52.3%), respectively (fig. 1). All studies used co-trimoxazole in standard doses for prophylaxis. Three publications also included co-amoxiclav or nitrofurantoin as alternative treatments and only 1 study was placebo controlled. In the high grade VUR group the risk of recurrent febrile UTI was 22.46% (53 of 236 patients) in those who received antibiotics vs 29.79% (70 of 235) in those who did not receive prophylaxis. The relative risk of treatment failure on antibiotic prophylaxis was 0.75 (95% CI 0.56e1.01) and the ARR of UTI was 7.33%. NNT was 13.64 (p ¼ 0.05, fig. 2). In children with low grade VUR the risk of febrile UTI was 6.10% (15 of 246) in those who received antibiotics vs 6.96% (16 of 230) in those who did not receive any treatment. The relative risk of treatment failure was 0.87 (95% CI 0.42e1.82) and the ARR of UTI was 0.86%. NNT was 117 (p ¼ 0.83, fig. 2). It was not possible to compare UTI rates in boys vs girls because of missing data in the studies included. When RIVUR data were included in the metaanalysis, the total number of patients increased to 1,593, comprising 61.89% of patients from the 6 previous studies and 38.11% from the RIVUR study.20 Of these patients 1,221 were girls (76.64%) and 372 were boys (23.36%), and 751 (47.29%) and 837 (52.71%) had dilated and nondilated VUR, respectively (see table). In this scenario in children with high grade VUR the risk of recurrent febrile UTI was 20.84% (79 of 379) in those who received antibiotics vs 29.03% (108 of 372) in those who did not receive
1774
ANTIBIOTIC PROPHYLAXIS FOR URINARY TRACT INFECTIONS IN VESICOURETERAL REFLUX
Figure 1. Recurrent febrile UTI/pyelonephritis in dilated and nondilated groups with vs without antimicrobial prophylaxis before RIVUR.20
prophylaxis. The relative risk of treatment failure was 0.72 (95% CI 0.56e0.92) and the ARR of UTI was 8.23%. NNT was 12.15 (p ¼ 0.008, see table). In patients with low grade VUR the risk of recurrent UTI was 6.44% (26 of 404) vs 12.94% (51 of 394) in those who did vs did not receive prophylaxis. The relative risk of treatment failure was 0.51 (95% CI 0.32e0.79) and the ARR of UTI was 6.51%. NNT was 15.36 (p ¼ 0.002, table).
DISCUSSION Antibiotic prophylaxis has been widely used to prevent UTIs in children with VUR despite the lack of strong high quality evidence to support its efficacy. Since the last decade, studies have questioned the benefit of prolonged low dose antibiotics to prevent UTIs in children with VUR. Few of those studies were randomized, controlled trials. Controversies have arisen on whether antibiotic prophylaxis is or is not beneficial in children with VUR and on the degree of VUR that requires prophylaxis. Important specialty societies such as the AAP have published guidelines5e8 based on aggregate data from those studies. The guidelines state that antimicrobial prophylaxis to prevent recurrent febrile UTIs in infants with grade I to IV VUR might not be necessary. Therefore, voiding cystourethrogram should not be routinely recommended after a first UTI. Our analysis done before RIVUR publication20 showed that children with nondilated VUR (grades
I and II) had a low benefit from being placed on antibiotic prophylaxis since it was necessary to treat 129 children to avoid 1 UTI. In contrast, those with a higher VUR grade (III and IV) would benefit from this prophylaxis because the number needed to treat was 13.6. The risk of recurrent febrile UTIs in patients with grades III and IV reflux was 22.5% compared to 5.31% in those with low grade reflux.20 In 2006 Garin et al reported one of the first studies to raise the question of the limited benefits of antibiotic prophylaxis.9 In that randomized, controlled trial 218 children completed the 1-year followup who had acute pyelonephritis with or without VUR up to grade III at ages between 3 months and 18 years. Children with VUR on prophylaxis were at greater risk for a new pyelonephritis episode than those not on prophylaxis (p ¼ 0.029) and most of those with recurrent infections had VUR grade III. This study included children with a wide age range from toddlers to older adolescents. They had characteristics that were not comparable, such as the presence or absence of VUR and the fact that controls received no treatment instead of placebo. Urine specimens were collected by different methods and different types of antibiotics were used. Montini et al studied 338 children with primary UTI with VUR up to grade III and without VUR.10 Bivariate analysis and the Cox proportional hazard model revealed that higher reflux grade and younger age of the child were associated with a greater risk of recurrent UTIs. NNT to prevent a single recurrence
ANTIBIOTIC PROPHYLAXIS FOR URINARY TRACT INFECTIONS IN VESICOURETERAL REFLUX
1775
Figure 2. Recurrent febrile UTI/pyelonephritis in dilated and nondilated groups with vs without antimicrobial prophylaxis, including RIVUR.20
of febrile UTI was 41.7 in 1 year. They found no difference in scar formation in the groups with vs without prophylaxis and concluded that prophylaxis does not decrease the rate of recurrent febrile UTIs during 12 months after the initial febrile UTI in children with or without primary nonsevere reflux. This study as well as that by Garin et al9 had no placebo group and used more than 1 type of antibiotic, and not all patients had VUR. Urine specimens were collected with urinary bags, which compromises the accuracy of confirming UTI. Pennesi et al investigated 100 patients 1 day to 30 months old after the first episode of pyelonephritis who had VUR grades II to IV.11 Patients were randomly divided by prophylaxis vs no prophylaxis and followed up to 4 years. The risk of at least 1 pyelonephritis recurrence was slightly higher but not statistically significant in the prophylaxis group
compared to controls (relative risk 1.2, 95% CI 0.68e2.11). They found no statistical difference in recurrent febrile UTIs among the different VUR grades. The study was not placebo controlled, used more than 1 type of antibiotic, excluded grade I reflux and used bags to collect urine. In addition, Roussey-Kesler et al reported a study in 225 patients 1 month to 3 years old with grade I to III VUR who were randomly assigned to prophylaxis vs no prophylaxis.12 There was no significant difference in febrile UTIs between the 2 groups. However, males were at greater risk for recurrent UTIs, most notably those with grade III VUR. Bags were used to collect urine and there was no placebo. The study that enrolled the greatest number of patients was done by Craig et al.13 The 576 patients were younger than 18 years (median age at study
RR, UTI ARR and NTT in dilated and nondilated groups before and after RIVUR study publication20 Dilated Before RIVUR: Relative risk (95% CI) % UTI ARR NNT After RIVUR: Relative risk (95% CI) % UTI ARR NNT
Nondilated
0.75 (0.56e1.01) 7.33 13.64
0.87 (0.42e1.82) 0.86 117
0.72 (0.56e0.92) 8.23 12.15
0.51 (0.32e0.79) 6.51 15.36
1776
ANTIBIOTIC PROPHYLAXIS FOR URINARY TRACT INFECTIONS IN VESICOURETERAL REFLUX
entry 14 months), had a history of a symptomatic UTI and were randomly assigned to prophylaxis or placebo. Patients with all VUR grades were included in analysis. At 12 months of followup Craig et al concluded that prophylaxis with trimethoprimsulfamethoxazole was associated with a decreased number of UTIs in predisposed children.13 This was the only placebo controlled study. In contrast to previous studies, they found a benefit of antibiotic prophylaxis to prevent UTIs. However, the study included patients with a wide age range (0 to 18 years) as well as patients with grade V reflux. Urine samples were obtained by different methods. The most recent study that fulfilled the inclusion criteria of our meta-analysis and was published before the RIVUR trial20 was the Swedish reflux study reported by Brandstr€ om et al in 2011.14 A total of 203 children younger than 2 years who had grade III or IV VUR were randomized to prophylaxis, endoscopic treatment or no treatment and followed for 2 years. This study as well as that by Craig et al13 demonstrated the positive effect of prophylaxis to reduce the rate of febrile UTIs and new renal damage, especially in girls. In contrast to the study by Roussey-Kesler et al,12 there was no benefit of prophylaxis in boys. The method of urine collection was not specified in the Swedish trial.14 None of the 6 randomized, controlled trials in our meta-analysis that were published before the RIVUR study20 evaluated family adhesion to prophylaxis.9e14 Therefore, it is uncertain whether patients in the prophylaxis group were receiving the prescribed antibiotic. In addition, while some discrepancies can be observed in the results when reviewing these studies, most of them show that higher grade VUR carries a greater risk of UTIs and such patients may benefit from prophylaxis. When we analyzed the data from these select studies and separated patients into groups with dilated and nondilated VUR, long-term, low dose antibiotics were beneficial to prevent recurrent febrile UTIs in patients with dilated VUR. However, in those with
low grade VUR no strong evidence supported the use of antibiotic prophylaxis to prevent new episodes of pyelonephritis. These data are in agreement with previously published guidelines.5e8 After aggregating the RIVUR data into our initial pooled analysis we observed that the RIVUR study20 accounted for more than a third of the patients in our analysis. Inclusion also changed the characteristics of the population in the meta-analysis, increasing the incidence of girls from 67.2% to 76.64%. After including the RIVUR patients we found that independent of VUR grade prophylactic antibiotics were helpful to prevent recurrent UTIs in children with VUR. We also noted that low vs high grade VUR carried a similar relative risk of treatment failure (0.51 vs 0.72), absolute risk reduction of febrile UTI (6.51% vs 8.23%) and NNT (15.36 vs 12.15). Although the RIVUR study was a well designed, placebo controlled trial,20 some issues merit discussion. The study population may not reflect most children seen in routine daily clinical practice. There was a great discrepancy between males and females with females representing 91.9% of the population while the literature shows a prevalence that is about 4 times greater in females. Almost half of the patients in the RIVUR study were younger than 11 months and VUR is more prevalent in males in that age group. These recent findings reinforce the idea that children with dilated VUR should benefit from antibiotic prophylaxis. In contrast to previous findings, it also suggests that even those with low grade VUR may need prophylaxis.
CONCLUSIONS VUR management is still controversial and has changed over the years. In contrast to recent published studies and guidelines, the results of this meta-analysis support antibiotic prophylaxis in all children with VUR regardless of reflux grade. More studies are needed to confirm these findings.
REFERENCES 1. American Academy of Pediatrics, Committee on Quality Improvement, Subcommittee on Urinary Tract Infection: Practice parameter: the diagnosis, treatment, and evaluation of the initial urinary tract infection in febrile infants and young children. Pediatrics 1999; 103: 843. 2. Wald E: Urinary tract infections in infants and children: a comprehensive overview. Curr Opin Pediatr 2004; 16: 85. 3. Gordon ZN, Caldamone A and Ellsworth P: Evaluation and management of vesicoureteral
reflux: a decade of change. Med Health R I 2010; 93: 342, 347. 4. Matoo TK: Evidence for and against urinary prophylaxis in vesicoureteral reflux. Pediatr Nephrol 2010; 25: 2379. 5. Roberts KB and Subcommittee on Urinary Tract Infection, Steering Committee on Quality Improvement and Management: Urinary tract infection: clinical practice guideline for the diagnosis and management of the initial UTI in febrile infants and children 2 to 24 months. Pediatrics 2011; 128: 595.
6. Finnell SM, Carroll AE, Downs SM; Subcommittee on Urinary Tract Infection: Technical report diagnosis and management of an initial UTI in febrile infants and young children. Pediatrics 2011; 128: e749. 7. Wan J, Skoog SJ, Hulbert WC et al; Executive Committee, Section on Urology, American Academy of Pediatrics: Section on Urology response to new Guidelines for the Diagnosis and Management of UTI. Pediatrics 2012; 129: e1051. 8. Mori R, Lakhanpaul M, Verrier KJ et al: Diagnosis and management of urinary tract infection
ANTIBIOTIC PROPHYLAXIS FOR URINARY TRACT INFECTIONS IN VESICOURETERAL REFLUX
in children: summary of NICE guidance. BMJ 2007; 335: 395.
from a prospective randomized study. J Urol 2008; 179: 674.
9. Garin EH, Olavarria F, Garcia Nieto V et al: Clinical significance of primary vesicoureteral reflux and urinary antibiotic prophylaxis after acute pyelonephritis: a multicenter, randomized, controlled study. Pediatrics 2006; 117: 626.
13. Craig JC, Simpson JM, Williams GJ et al: Antibiotic prophylaxis and recurrent urinary tract infection in children. N Engl J Med 2009; 361: 1748.
10. Montini G, Rigon L, Zucchetta P et al: Prophylaxis after first febrile urinary tract infection in children? A multicenter, randomized, controlled, noninferiority trial. Pediatrics 2008; 122: 1064. 11. Pennesi MTL, Peratoner L, Bordugo A et al: Is antibiotic prophylaxis in children with vesicoureteral reflux effective in preventing pyelonephritis and renal scars? A randomized, controlled trial. Pediatrics 2008; 121: e1489. 12. Roussey-Kesler G, Gadjos V, Idres N et al: Antibiotic prophylaxis for the prevention of recurrent urinary tract infection in children with low grade vesicoureteral reflux: results
14. Brandstr€om P, Jodal U, Sillen U et al: The Swedish reflux trial: review of a randomized, controlled trial in children with dilating vesicoureteral reflux. J Pediatr Urol 2011; 7: 594. 15. Mathew JL: Antibiotic prophylaxis following urinary tract infection in children: a systematic review of randomized controlled trials. Indian Pediatr 2010; 47: 599. 16. Ammenti A, Cataldi L, Chimenz R et al; Italian Society of Pediatric Nephrology: Febrile urinary tract infections in young children: recommendations for the diagnosis, treatment and follow-up. Acta Paediatr 2012; 101: 451.
1777
17. Hoberman A and Keren R: Antimicrobial prophylaxis for urinary tract infection in children. N Engl J Med 2009; 361: 1804. 18. Mantadakis E, Vouloumanou EK, Georgantzi GG et al: Acute Tc-99m DMSA scan for identifying dilating vesicoureteral reflux in children: a metaanalysis. Pediatrics 2011; 129: e169. 19. Williams G and Craig JC: Long-term antibiotics for preventing recurrent urinary tract infection in children. Cochrane Database Syst Rev 2011; 3: CD001534. 20. RIVUR Trial Investigators: Antimicrobial prophylaxis for children with vesicoureteral reflux. N Engl J Med 2014; 370: 2367. 21. Jadad AR, Moore RA, Carroll D et al: Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996; 17: 1.
EDITORIAL COMMENT This is the second meta-analysis including the newer RIVUR data with conclusions contradictory to those obtained by the AAP in 2011 (references 5 and 20 in article).1 This is despite the fact that there are significant flaws in most of these metaanalyses, including lack of blinding, the acceptance of bagged urine specimens from nontoilet trained infants, reliance on renal ultrasound to assess for scarring, failure to assess medication adherence and reliance on unsubstantiated local readings of radiographs, among others. Heterogeneity of populations is an additional factor since there are many more uncircumcised males in studies from Europe and South America, and adolescents were included on occasion. One might reasonably ask whether data from some of these series have any credibility at all. The fallback argument against CAP is that while infections are fewer, renal scarring rates are not reduced, although the Swedish reflux trial showed a clear difference.2 Pediatric nephrologists are most concerned with the renal impact of VUR and may not always appreciate the morbidity of recurrent urinary infections even in the absence of pyelonephritic scarring. Recurrent cystitis with or without fever is uncomfortable, requires medical intervention and is of considerable concern to parents. Furthermore, new renal scarring can only
develop as a consequence of recurrent UTIs. Especially notable is the fact that the RIVUR trial demonstrated the largest impact of CAP for decreasing infection in children with grades I and II reflux. While the RIVUR trial was the largest of its kind, it was not powered to show significant scarring differences (reference 20 in article). It was estimated that 8,000 subjects would be needed for such an analysis. In addition, the short observation period of 2 years and the fact that families were repetitively reminded of the signs and symptoms of infection and obtained rapid treatment via RIVUR nurse coordinators decreased the likelihood of renal damage. RIVUR data alone as well as the combined data in this meta-analysis should prompt a reevaluation of the AAP recommendation that VUR need not be identified after the first infection (reference 5 in article). The greatest challenge remains identifying the minority of children with VUR who are at risk for renal damage. Until this is reliably achieved CAP along with surgical intervention when warranted remain essential modes of treatment. Saul P. Greenfield Division of Pediatric Urology Women and Children’s Hospital of Buffalo Buffalo, New York
REFERENCES 1. Wang HHS, Gbadegesin RA, Foreman JW et al: Efficacy of antibiotic prophylaxis in children with vesicoureteral reflux: systematic review and meta-analysis. J Urol 2015; 193: 963. 2. Brandstr€om P, Neveus T, Sixt R et al: The Swedish reflux trial in children: IV. Renal damage. J Urol 2010; 184: 292.
Abbreviations and Acronyms AAP ¼ American Academy of Pediatrics ARR ¼ absolute risk reduction CAP ¼ continuous antibiotic prophylaxis NNT ¼ number needed to treat RIVUR ¼ Randomized Intervention for Children with Vesicoureteral Reflux UTI ¼ urinary tract infection VUR ¼ vesicoureteral reflux Accepted for publication October 10, 2014. * Correspondence: Universidade Estadual de Feira de Santana, Feira de Santana, Brazil (e-mail: [email protected]).
Purpose: The followup and treatment of children with vesicoureteral reflux has been debated for many years. Antibiotic prophylaxis has a role for preventing urinary tract infection in these children. Recent studies and guidelines suggested that prophylaxis has little or no role in preventing urinary tract infection in those children, especially those with low grades (I and II) of reflux. Materials and Methods: We analyzed all published randomized, controlled trials comparing antibiotic prophylaxis vs no prophylaxis or placebo in children with vesicoureteral reflux. The children were divided into those with nondilated (grades I and II) and dilated (grades III and IV) vesicoureteral reflux. After data were analyzed the RIVUR study was published and, therefore, it was added to the analyzed data. Results: After analyzing the first published studies we found that antibiotic prophylaxis would be beneficial only in children with high grade vesicoureteral reflux. With the addition of the data in the RIVUR study these results changed. The new pooled data support antibiotic prophylaxis in all children with vesicoureteral reflux. Conclusions: Vesicoureteral reflux management is still controversial. In contrast to recently published studies and guidelines, this meta-analysis supports antibiotic prophylaxis in all children with vesicoureteral reflux regardless of reflux grade. More studies are needed to support this finding. Key Words: kidney, urinary tract infections, vesico-ureteral reflux, antibiotic prophylaxis, pyelonephritis
LOW dose, prolonged antibiotic prophylaxis has been used for more than 4 decades to prevent recurrent UTIs and consequent renal scars in infants and children with VUR.1,2 This practice was based on expert opinions
1772
j
www.jurology.com
and poorly designed, nonrandomized studies, and supported by many professional societies.3e8 More recently better designed studies questioned the real benefits of antibiotic prophylaxis in children with VUR and showed a
0022-5347/15/1935-1772/0 THE JOURNAL OF UROLOGY® © 2015 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC.
http://dx.doi.org/10.1016/j.juro.2014.10.092 Vol. 193, 1772-1777, May 2015 Printed in U.S.A.
ANTIBIOTIC PROPHYLAXIS FOR URINARY TRACT INFECTIONS IN VESICOURETERAL REFLUX
lack of convincing evidence for this practice, adding to the current controversy on this topic.9e17 These studies and others raised many questions on whether children who receive antibiotic prophylaxis are better protected from new onset pyelonephritis, which children would benefit the most from prophylaxis and whether prophylaxis really protects the kidneys from new scar formation. Therefore, other meta-analyses and systematic reviews, including a Cochrane commentary, were done with the objective of better understanding all published data on this subject.4,18,19 These studies demonstrated that many of the mentioned questions remain to be answered and strong evidence supporting this practice is still lacking. However, no previously published meta-analysis analyzed data separating high grade (dilated, grades III and IV) from low grade (nondilated, grades I and II) VUR. Recently the multicenter, randomized RIVUR trial showed that antibiotic prophylaxis was effective in decreasing recurrent UTIs in children with VUR.20 This suggested that the AAP and NICE (National Institute for Health and Care Excellence) recommendations5e8 should be reevaluated. We systematically reviewed the VUR literature to evaluate the effects of CAP to prevent recurrent UTIs in children with dilating VUR (grades III and IV) compared to those with no dilating VUR (grades I and II). We hypothesized that antibiotic prophylaxis would be effective for decreasing the number of recurrent febrile UTIs compared to no treatment in children with grades III and V VUR but not in those with grades I and II.
MATERIAL AND METHODS On November 30, 2013 we performed an electronic survey of VUR trials published in English using MEDLINEÒ, EmbaseÒ and ScopusÒ. We included all randomized and quasi-randomized, controlled trials comparing antibiotics to placebo/no treatment or comparing 2 or more antibiotics administered daily for at least 2 months to prevent recurrent UTIs in children with VUR. Studies of curative treatment, lifestyle interventions and preoperative antibiotic prophylaxis were excluded from analysis. The primary outcome was a recurrent febrile UTI. The initial review and first update were done by 3 of us (FCdCM, EFM and MCB). All selected articles were checked independently against study inclusion criteria and design by 2 groups of 3 reviewers (FCdCM, EFM and MCB or RBT, VPP and ZMS). In case of discordance other experts were consulted (JdB or JMBN). Data extraction from eligible studies was performed independently by the same reviewers with discrepancies in data extraction solved by a third referee (LHB). The internal validity of included trials was assessed by 2 reviewers using the Jadad scale.21 Our initial analysis comprised 6 articles that met study inclusion criteria. We collected data on all children in
1773
these studies who were reported to have VUR. Nondilated VUR was defined as low grade (I and II) and dilated VUR was defined as higher grade (III to IV).18 Secondary analysis was performed to compare pooled results before RIVUR publication to that of patients in RIVUR.20 A random effects model was used to estimate the risk ratio and risk difference for recurrent febrile UTIs with the 95% CI. Relative risk was calculated as the risk ratio of exposed to nonexposed, that is exposure to prophylactic antibiotics. NNT was calculated to synthesize the potential beneficial role of antibiotic prophylaxis. I2 was used to quantify heterogeneity. Meta-analysis was performed when deemed appropriate, ie when heterogeneity among studies was not considered too high. All statistical analysis was done with Comprehensive Meta-Analysis (Biostat, Englewood, New Jersey).
RESULTS Pooled data from the 6 randomized, controlled trials published until April 2014 included a total of 986 children, of whom 663 (67.2%) were female and 323 (32.8%) were male.9e14 Dilated and nondilated VUR was observed in 471 (47.7%) and 515 patients (52.3%), respectively (fig. 1). All studies used co-trimoxazole in standard doses for prophylaxis. Three publications also included co-amoxiclav or nitrofurantoin as alternative treatments and only 1 study was placebo controlled. In the high grade VUR group the risk of recurrent febrile UTI was 22.46% (53 of 236 patients) in those who received antibiotics vs 29.79% (70 of 235) in those who did not receive prophylaxis. The relative risk of treatment failure on antibiotic prophylaxis was 0.75 (95% CI 0.56e1.01) and the ARR of UTI was 7.33%. NNT was 13.64 (p ¼ 0.05, fig. 2). In children with low grade VUR the risk of febrile UTI was 6.10% (15 of 246) in those who received antibiotics vs 6.96% (16 of 230) in those who did not receive any treatment. The relative risk of treatment failure was 0.87 (95% CI 0.42e1.82) and the ARR of UTI was 0.86%. NNT was 117 (p ¼ 0.83, fig. 2). It was not possible to compare UTI rates in boys vs girls because of missing data in the studies included. When RIVUR data were included in the metaanalysis, the total number of patients increased to 1,593, comprising 61.89% of patients from the 6 previous studies and 38.11% from the RIVUR study.20 Of these patients 1,221 were girls (76.64%) and 372 were boys (23.36%), and 751 (47.29%) and 837 (52.71%) had dilated and nondilated VUR, respectively (see table). In this scenario in children with high grade VUR the risk of recurrent febrile UTI was 20.84% (79 of 379) in those who received antibiotics vs 29.03% (108 of 372) in those who did not receive
1774
ANTIBIOTIC PROPHYLAXIS FOR URINARY TRACT INFECTIONS IN VESICOURETERAL REFLUX
Figure 1. Recurrent febrile UTI/pyelonephritis in dilated and nondilated groups with vs without antimicrobial prophylaxis before RIVUR.20
prophylaxis. The relative risk of treatment failure was 0.72 (95% CI 0.56e0.92) and the ARR of UTI was 8.23%. NNT was 12.15 (p ¼ 0.008, see table). In patients with low grade VUR the risk of recurrent UTI was 6.44% (26 of 404) vs 12.94% (51 of 394) in those who did vs did not receive prophylaxis. The relative risk of treatment failure was 0.51 (95% CI 0.32e0.79) and the ARR of UTI was 6.51%. NNT was 15.36 (p ¼ 0.002, table).
DISCUSSION Antibiotic prophylaxis has been widely used to prevent UTIs in children with VUR despite the lack of strong high quality evidence to support its efficacy. Since the last decade, studies have questioned the benefit of prolonged low dose antibiotics to prevent UTIs in children with VUR. Few of those studies were randomized, controlled trials. Controversies have arisen on whether antibiotic prophylaxis is or is not beneficial in children with VUR and on the degree of VUR that requires prophylaxis. Important specialty societies such as the AAP have published guidelines5e8 based on aggregate data from those studies. The guidelines state that antimicrobial prophylaxis to prevent recurrent febrile UTIs in infants with grade I to IV VUR might not be necessary. Therefore, voiding cystourethrogram should not be routinely recommended after a first UTI. Our analysis done before RIVUR publication20 showed that children with nondilated VUR (grades
I and II) had a low benefit from being placed on antibiotic prophylaxis since it was necessary to treat 129 children to avoid 1 UTI. In contrast, those with a higher VUR grade (III and IV) would benefit from this prophylaxis because the number needed to treat was 13.6. The risk of recurrent febrile UTIs in patients with grades III and IV reflux was 22.5% compared to 5.31% in those with low grade reflux.20 In 2006 Garin et al reported one of the first studies to raise the question of the limited benefits of antibiotic prophylaxis.9 In that randomized, controlled trial 218 children completed the 1-year followup who had acute pyelonephritis with or without VUR up to grade III at ages between 3 months and 18 years. Children with VUR on prophylaxis were at greater risk for a new pyelonephritis episode than those not on prophylaxis (p ¼ 0.029) and most of those with recurrent infections had VUR grade III. This study included children with a wide age range from toddlers to older adolescents. They had characteristics that were not comparable, such as the presence or absence of VUR and the fact that controls received no treatment instead of placebo. Urine specimens were collected by different methods and different types of antibiotics were used. Montini et al studied 338 children with primary UTI with VUR up to grade III and without VUR.10 Bivariate analysis and the Cox proportional hazard model revealed that higher reflux grade and younger age of the child were associated with a greater risk of recurrent UTIs. NNT to prevent a single recurrence
ANTIBIOTIC PROPHYLAXIS FOR URINARY TRACT INFECTIONS IN VESICOURETERAL REFLUX
1775
Figure 2. Recurrent febrile UTI/pyelonephritis in dilated and nondilated groups with vs without antimicrobial prophylaxis, including RIVUR.20
of febrile UTI was 41.7 in 1 year. They found no difference in scar formation in the groups with vs without prophylaxis and concluded that prophylaxis does not decrease the rate of recurrent febrile UTIs during 12 months after the initial febrile UTI in children with or without primary nonsevere reflux. This study as well as that by Garin et al9 had no placebo group and used more than 1 type of antibiotic, and not all patients had VUR. Urine specimens were collected with urinary bags, which compromises the accuracy of confirming UTI. Pennesi et al investigated 100 patients 1 day to 30 months old after the first episode of pyelonephritis who had VUR grades II to IV.11 Patients were randomly divided by prophylaxis vs no prophylaxis and followed up to 4 years. The risk of at least 1 pyelonephritis recurrence was slightly higher but not statistically significant in the prophylaxis group
compared to controls (relative risk 1.2, 95% CI 0.68e2.11). They found no statistical difference in recurrent febrile UTIs among the different VUR grades. The study was not placebo controlled, used more than 1 type of antibiotic, excluded grade I reflux and used bags to collect urine. In addition, Roussey-Kesler et al reported a study in 225 patients 1 month to 3 years old with grade I to III VUR who were randomly assigned to prophylaxis vs no prophylaxis.12 There was no significant difference in febrile UTIs between the 2 groups. However, males were at greater risk for recurrent UTIs, most notably those with grade III VUR. Bags were used to collect urine and there was no placebo. The study that enrolled the greatest number of patients was done by Craig et al.13 The 576 patients were younger than 18 years (median age at study
RR, UTI ARR and NTT in dilated and nondilated groups before and after RIVUR study publication20 Dilated Before RIVUR: Relative risk (95% CI) % UTI ARR NNT After RIVUR: Relative risk (95% CI) % UTI ARR NNT
Nondilated
0.75 (0.56e1.01) 7.33 13.64
0.87 (0.42e1.82) 0.86 117
0.72 (0.56e0.92) 8.23 12.15
0.51 (0.32e0.79) 6.51 15.36
1776
ANTIBIOTIC PROPHYLAXIS FOR URINARY TRACT INFECTIONS IN VESICOURETERAL REFLUX
entry 14 months), had a history of a symptomatic UTI and were randomly assigned to prophylaxis or placebo. Patients with all VUR grades were included in analysis. At 12 months of followup Craig et al concluded that prophylaxis with trimethoprimsulfamethoxazole was associated with a decreased number of UTIs in predisposed children.13 This was the only placebo controlled study. In contrast to previous studies, they found a benefit of antibiotic prophylaxis to prevent UTIs. However, the study included patients with a wide age range (0 to 18 years) as well as patients with grade V reflux. Urine samples were obtained by different methods. The most recent study that fulfilled the inclusion criteria of our meta-analysis and was published before the RIVUR trial20 was the Swedish reflux study reported by Brandstr€ om et al in 2011.14 A total of 203 children younger than 2 years who had grade III or IV VUR were randomized to prophylaxis, endoscopic treatment or no treatment and followed for 2 years. This study as well as that by Craig et al13 demonstrated the positive effect of prophylaxis to reduce the rate of febrile UTIs and new renal damage, especially in girls. In contrast to the study by Roussey-Kesler et al,12 there was no benefit of prophylaxis in boys. The method of urine collection was not specified in the Swedish trial.14 None of the 6 randomized, controlled trials in our meta-analysis that were published before the RIVUR study20 evaluated family adhesion to prophylaxis.9e14 Therefore, it is uncertain whether patients in the prophylaxis group were receiving the prescribed antibiotic. In addition, while some discrepancies can be observed in the results when reviewing these studies, most of them show that higher grade VUR carries a greater risk of UTIs and such patients may benefit from prophylaxis. When we analyzed the data from these select studies and separated patients into groups with dilated and nondilated VUR, long-term, low dose antibiotics were beneficial to prevent recurrent febrile UTIs in patients with dilated VUR. However, in those with
low grade VUR no strong evidence supported the use of antibiotic prophylaxis to prevent new episodes of pyelonephritis. These data are in agreement with previously published guidelines.5e8 After aggregating the RIVUR data into our initial pooled analysis we observed that the RIVUR study20 accounted for more than a third of the patients in our analysis. Inclusion also changed the characteristics of the population in the meta-analysis, increasing the incidence of girls from 67.2% to 76.64%. After including the RIVUR patients we found that independent of VUR grade prophylactic antibiotics were helpful to prevent recurrent UTIs in children with VUR. We also noted that low vs high grade VUR carried a similar relative risk of treatment failure (0.51 vs 0.72), absolute risk reduction of febrile UTI (6.51% vs 8.23%) and NNT (15.36 vs 12.15). Although the RIVUR study was a well designed, placebo controlled trial,20 some issues merit discussion. The study population may not reflect most children seen in routine daily clinical practice. There was a great discrepancy between males and females with females representing 91.9% of the population while the literature shows a prevalence that is about 4 times greater in females. Almost half of the patients in the RIVUR study were younger than 11 months and VUR is more prevalent in males in that age group. These recent findings reinforce the idea that children with dilated VUR should benefit from antibiotic prophylaxis. In contrast to previous findings, it also suggests that even those with low grade VUR may need prophylaxis.
CONCLUSIONS VUR management is still controversial and has changed over the years. In contrast to recent published studies and guidelines, the results of this meta-analysis support antibiotic prophylaxis in all children with VUR regardless of reflux grade. More studies are needed to confirm these findings.
REFERENCES 1. American Academy of Pediatrics, Committee on Quality Improvement, Subcommittee on Urinary Tract Infection: Practice parameter: the diagnosis, treatment, and evaluation of the initial urinary tract infection in febrile infants and young children. Pediatrics 1999; 103: 843. 2. Wald E: Urinary tract infections in infants and children: a comprehensive overview. Curr Opin Pediatr 2004; 16: 85. 3. Gordon ZN, Caldamone A and Ellsworth P: Evaluation and management of vesicoureteral
reflux: a decade of change. Med Health R I 2010; 93: 342, 347. 4. Matoo TK: Evidence for and against urinary prophylaxis in vesicoureteral reflux. Pediatr Nephrol 2010; 25: 2379. 5. Roberts KB and Subcommittee on Urinary Tract Infection, Steering Committee on Quality Improvement and Management: Urinary tract infection: clinical practice guideline for the diagnosis and management of the initial UTI in febrile infants and children 2 to 24 months. Pediatrics 2011; 128: 595.
6. Finnell SM, Carroll AE, Downs SM; Subcommittee on Urinary Tract Infection: Technical report diagnosis and management of an initial UTI in febrile infants and young children. Pediatrics 2011; 128: e749. 7. Wan J, Skoog SJ, Hulbert WC et al; Executive Committee, Section on Urology, American Academy of Pediatrics: Section on Urology response to new Guidelines for the Diagnosis and Management of UTI. Pediatrics 2012; 129: e1051. 8. Mori R, Lakhanpaul M, Verrier KJ et al: Diagnosis and management of urinary tract infection
ANTIBIOTIC PROPHYLAXIS FOR URINARY TRACT INFECTIONS IN VESICOURETERAL REFLUX
in children: summary of NICE guidance. BMJ 2007; 335: 395.
from a prospective randomized study. J Urol 2008; 179: 674.
9. Garin EH, Olavarria F, Garcia Nieto V et al: Clinical significance of primary vesicoureteral reflux and urinary antibiotic prophylaxis after acute pyelonephritis: a multicenter, randomized, controlled study. Pediatrics 2006; 117: 626.
13. Craig JC, Simpson JM, Williams GJ et al: Antibiotic prophylaxis and recurrent urinary tract infection in children. N Engl J Med 2009; 361: 1748.
10. Montini G, Rigon L, Zucchetta P et al: Prophylaxis after first febrile urinary tract infection in children? A multicenter, randomized, controlled, noninferiority trial. Pediatrics 2008; 122: 1064. 11. Pennesi MTL, Peratoner L, Bordugo A et al: Is antibiotic prophylaxis in children with vesicoureteral reflux effective in preventing pyelonephritis and renal scars? A randomized, controlled trial. Pediatrics 2008; 121: e1489. 12. Roussey-Kesler G, Gadjos V, Idres N et al: Antibiotic prophylaxis for the prevention of recurrent urinary tract infection in children with low grade vesicoureteral reflux: results
14. Brandstr€om P, Jodal U, Sillen U et al: The Swedish reflux trial: review of a randomized, controlled trial in children with dilating vesicoureteral reflux. J Pediatr Urol 2011; 7: 594. 15. Mathew JL: Antibiotic prophylaxis following urinary tract infection in children: a systematic review of randomized controlled trials. Indian Pediatr 2010; 47: 599. 16. Ammenti A, Cataldi L, Chimenz R et al; Italian Society of Pediatric Nephrology: Febrile urinary tract infections in young children: recommendations for the diagnosis, treatment and follow-up. Acta Paediatr 2012; 101: 451.
1777
17. Hoberman A and Keren R: Antimicrobial prophylaxis for urinary tract infection in children. N Engl J Med 2009; 361: 1804. 18. Mantadakis E, Vouloumanou EK, Georgantzi GG et al: Acute Tc-99m DMSA scan for identifying dilating vesicoureteral reflux in children: a metaanalysis. Pediatrics 2011; 129: e169. 19. Williams G and Craig JC: Long-term antibiotics for preventing recurrent urinary tract infection in children. Cochrane Database Syst Rev 2011; 3: CD001534. 20. RIVUR Trial Investigators: Antimicrobial prophylaxis for children with vesicoureteral reflux. N Engl J Med 2014; 370: 2367. 21. Jadad AR, Moore RA, Carroll D et al: Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996; 17: 1.
EDITORIAL COMMENT This is the second meta-analysis including the newer RIVUR data with conclusions contradictory to those obtained by the AAP in 2011 (references 5 and 20 in article).1 This is despite the fact that there are significant flaws in most of these metaanalyses, including lack of blinding, the acceptance of bagged urine specimens from nontoilet trained infants, reliance on renal ultrasound to assess for scarring, failure to assess medication adherence and reliance on unsubstantiated local readings of radiographs, among others. Heterogeneity of populations is an additional factor since there are many more uncircumcised males in studies from Europe and South America, and adolescents were included on occasion. One might reasonably ask whether data from some of these series have any credibility at all. The fallback argument against CAP is that while infections are fewer, renal scarring rates are not reduced, although the Swedish reflux trial showed a clear difference.2 Pediatric nephrologists are most concerned with the renal impact of VUR and may not always appreciate the morbidity of recurrent urinary infections even in the absence of pyelonephritic scarring. Recurrent cystitis with or without fever is uncomfortable, requires medical intervention and is of considerable concern to parents. Furthermore, new renal scarring can only
develop as a consequence of recurrent UTIs. Especially notable is the fact that the RIVUR trial demonstrated the largest impact of CAP for decreasing infection in children with grades I and II reflux. While the RIVUR trial was the largest of its kind, it was not powered to show significant scarring differences (reference 20 in article). It was estimated that 8,000 subjects would be needed for such an analysis. In addition, the short observation period of 2 years and the fact that families were repetitively reminded of the signs and symptoms of infection and obtained rapid treatment via RIVUR nurse coordinators decreased the likelihood of renal damage. RIVUR data alone as well as the combined data in this meta-analysis should prompt a reevaluation of the AAP recommendation that VUR need not be identified after the first infection (reference 5 in article). The greatest challenge remains identifying the minority of children with VUR who are at risk for renal damage. Until this is reliably achieved CAP along with surgical intervention when warranted remain essential modes of treatment. Saul P. Greenfield Division of Pediatric Urology Women and Children’s Hospital of Buffalo Buffalo, New York
REFERENCES 1. Wang HHS, Gbadegesin RA, Foreman JW et al: Efficacy of antibiotic prophylaxis in children with vesicoureteral reflux: systematic review and meta-analysis. J Urol 2015; 193: 963. 2. Brandstr€om P, Neveus T, Sixt R et al: The Swedish reflux trial in children: IV. Renal damage. J Urol 2010; 184: 292.
