This review assessed treatments for vesicoureteric reflux. The authors concluded that it was unclear whether any intervention is effective in children with vesicoureteric reflux. The authors' conclusions are likely to be reliable.

The objectives were to assess the effect of treatment for vesicoureteric reflux (VUR) and to compare treatment with no treatment and with other treatments.

MEDLINE (from 1966 to February 2003) and EMBASE (from 1988 to February 2003) were searched for studies published in any language; some details of the search strategy were given. The Cochrane Controlled Trials Register was also searched and the reference lists of identified studies were checked. Researchers in the field were contacted for additional studies.

Study designs of evaluations included in the review

Randomised controlled trials (RCTs) or quasi-RCTs were eligible for inclusion.

Specific interventions included in the review

Studies of any type of treatment for VUR were eligible for inclusion. The included studies compared long-term (1 to 24 months) antibiotics with antibiotics plus surgery (ureteric reimplantation or Dx/HA copolymer subureteric implantation), or compared different antibiotic regimes (daily or intermittent antibiotics) with no treatment. The studies used a variety of open surgical techniques. The studies used the antibiotics trimethoprim (with or without sulphamethoxazole) or nitrofurantoin.

Participants included in the review

Studies of patients of any age with primary VUR diagnosed by micturating cystourethrogram, who did not have any major urological or structural abnormality, were eligible for inclusion. All of the included studies were of children aged younger than 15 years. Most of the included children had higher grades of VUR.

Outcomes assessed in the review

Studies that assessed urinary tract infection (UTI) or renal parenchymal injury were eligible for inclusion. UTI was defined in the individual studies (where stated) as greater than 100,000 colony forming units per mL. The review assessed symptomatic UTI, febrile UTI, new and progressive renal parenchymal abnormality (by using intravenous pyelography with or without Tm-dimercaptosuccinic acid scintigraphy), renal growth (by measuring changes in renal length standard deviation scores or renal area on intravenous pyelogram), end-stage renal failure, hypertension, glomerular filtration rate, resolution of reflux and adverse effects. The included studies assessed outcomes after 1 to 10 years.

How were decisions on the relevance of primary studies made?

Two reviewers independently reviewed the abstracts of identified studies and selected studies according to the inclusion criteria. Any disagreements were resolved through discussion with a third author. Additional information was sought from the authors of studies where eligibility was uncertain.

Validity was assessed, based on the following: the method of randomisation, allocation concealment, standardisation and blinding of the outcome assessment, intention-to-treat analysis, and losses to follow-up (see Other Publications of Related Interest). Three reviewers independently assessed validity using a standardised checklist. Any disagreements were resolved through discussion with a fourth author.

Three reviewers independently extracted data using a standardised checklist for the outcomes. Any disagreements were resolved through discussion with a fourth author. Data were extracted on the sample size, characteristics of the participants and interventions, inclusion criteria, treatment duration and outcomes. Only the most complete data set was extracted for each study. The relative risk (RR) and 95% confidence intervals (CIs) were extracted for dichotomous data from each study.

How were the studies combined?

The characteristics of the included studies were summarised in the text of the review. Studies that compared antibiotics with antibiotics plus surgery were combined using a meta-analysis. The pooled RR and 95% CI were calculated for UTI and renal parenchymal abnormality outcomes using a random-effects model. Data for the other outcomes were combined in a narrative.

How were differences between studies investigated?

Statistical heterogeneity was tested using Cochran's Q statistic, taking a P-value of less than 0.1 to indicate significant heterogeneity. Forest plots were presented for the meta-analyses of UTI outcomes and renal parenchymal abnormalities. The pooled RR and 95% CI were also calculated using a fixed-effect model. However, since the results were similar to the random-effects model, only the results from the random-effects model were presented in the review.

Eight RCTs were included (947 children enrolled; data were available for 859 children).

Five RCTs reported an adequate method of randomisation. Two RCTs reported that the radiologists were blinded to the treatment group. Three RCTs did not analyse data on an intention-to-treat basis and it was unclear whether any of the other RCTs has used intention-to-treat analysis. Studies reported low rates of losses to follow-up: 0 to 2% at 1 to 2 years, and 9 to 42% after 4 to 10 years.

Antibiotics versus antibiotics plus surgery (7 RCTs).

UTI: the meta-analysis showed no significant difference in UTI after 2 years or after 5 years between antibiotics alone and antibiotics plus surgery; the RR was 1.07 (95% CI: 0.55, 2.09) at 2 years (4 RCTs), and 0.99 (95% CI: 0.79, 1.26) after 5 years (3 RCTs). The meta-analysis showed that combined treatment reduced febrile UTI at 5 years compared with antibiotics alone; the RR (2 RCTs) was 0.43 (95% CI: 0.27, 0.70). No significant heterogeneity was detected for any of these meta-analyses (P>0.2). There was no significant difference in symptomatic UTI; the RR (European arm of 1 RCT) was 0.96 (95% CI: 0.67, 1.39).

Renal parenchymal abnormality: the meta-analysis showed no significant difference for new or progressive parenchymal abnormalities (detected using intravenous pyelography) at 2 years, or at 4 to 5 years. The RR of new defects was 1.06 (95% CI: 0.33, 3.42) at 2 years (2 RCTs) and 1.06 (95% CI: 0.77, 1.45) at 4 to 5 years (4 RCTs); the RR of progressive defects was 1.62 (95% CI: 0.25, 10.48) at 2 years (2 RCTs) and 0.97 (95% CI: 0.67, 1.40) at 4 to 5 years (3 RCTs). No significant heterogeneity was detected for any of these meta-analyses (P>0.1).

One RCT that also assessed renal parenchymal abnormalities using Tm-dimercaptosuccinic acid scintigraphy found that combined treatment increased new or progressive lesions compared with antibiotics alone, but the increase was not statistically significant (RR 1.05, 95% CI: 0.62, 1.77). Renal growth: it was not possible to perform a meta-analysis for studies reporting renal growth because of differences in the reported data. Four RCTs found no significant difference between treatments at 2 to 10 years.

Other outcomes.

End-stage renal failure (2 RCTs): three children in each treatment arm developed end-stage renal failure.

Hypertension (2 RCTs): five children treated with antibiotics alone developed hypertension compared with three who received combined treatment.

Glomerular filtration rate (4 RCTs): no individual study found any significant difference between the treatments.

Resolution of reflux (6 RCTs): the studies reported the results in different units (ureters or patients). Four RCTs found the resolution rates at 4 to 5 years were 93 to 99% after surgery and 16 to 49% for spontaneous resolution. One RCT found the resolution rate at 12 months was 69% after Dx/HA copolymer subureteric implantation compared with 38% with antibiotics alone.

Adverse effects: these were generally poorly reported. One arm of one RCT found postoperative urinary tract obstruction in 6.6% (10 out of 151) of children. One RCT (179 children) found no cases of obstruction after 5 years. None of the other studies mentioned obstruction.

Antibiotics compared with no treatment (1 RCT, 43 children entered, 29 analysed).

The RCT found no significant difference in UTI or renal parenchymal abnormality between treatments; the RR was 0.25 (95% CI: 0.03, 1.85) for UTI and 0.40 (95% CI: 0.02, 9.18) for renal parenchymal abnormality. The RCT might have been too small to detect a significant difference. The RCT did not report adverse effects.

It was unclear whether any intervention is effective in children with VUR.

The review question was clear in terms of the study design, intervention, participants and outcomes. Several relevant sources were searched and no language restrictions were applied. Since the search strategy was not reported in full, it could neither be evaluated nor replicated. More than one reviewer independently selected the studies, assessed validity and extracted the data; this reduces the potential for bias and errors. Validity was assessed using defined criteria and the results of the assessment were reported. Some relevant information on the included studies was tabulated. The data were appropriately combined in a meta-analysis and statistical heterogeneity was assessed. The evidence presented appears to support the authors' conclusions.

Practice: The authors stated that paediatricians and general practitioners should be aware that there is no good evidence on the effect of interventions for children with VUR. This needs to be considered when making decisions about the investigation of children with UTIs.

Research: The authors stated that well-designed, adequately powered RCTs are required to compare antibiotics with placebo in children with VUR.

Australian Kidney Foundation, grant number S2/99.

Schulz KF, Chambers I, Hoyes RJ, Altman DG. Empirical evidence of bias. Dimensions of methodological quality associated with estimates of treatment effects in controlled trials. JAMA 1995;273:408-12.

This additional published commentary may also be of interest. McTaggart S. Review: the only effect for surgery plus antiobiotics in vesicoureteric reflux is fewer febrile urinary tract infections. Evid Based Med 2004;9:41.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.