Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Circumcision for UTI

Systematic review
Results
Comment

Circumcision is the most common surgery carried out on children, for various reasons. Rates vary because of cultural and religious differences. Circumcision rates are as high as two-thirds in North America, 10-20% in Australia, but much lower in Europe and much of Asia. Circumcision is recommended for true phimosis where the foreskin cannot be drawn back to uncover the glans penis, or inflammation of the glans penis, or for recurrent urinary tract infection (UTI). How effective circumcision is in reducing UTI a systematic review [1] tells us.

Systematic review

The review sought studies of any design to the end of 2002, not limited by language, in several electronic databases and bibliographies. Studies with information on the effect of male circumcision on UTI were included if they provided enough information to be able to compare odds of UTI in a circumcised group with the odds in an uncircumcised group. There was no age restriction.

Results

Twelve studies fulfilled the inclusion criteria. Most examined infants, though one included adults and boys older than one year. One randomised trial had 70 patients, four cohort studies 400,000 patients, and seven case control studies 2,150 patients (Table 1).



Table 1: Effect of circumcision on UTI rates in studies of different design



Number of
UTI (%) in
Study type
Studies
Patients
Circumcised
Uncircumcised
Relative risk
(95% CI)
NNT
(95% CI)
Randomised trial
1
70
0.00
10
0.14 (0.01 to 2.6)
Cohort
4
400700
0.13
1.1
0.13 (0.12 to 0.15)
99 (93 to 106)
Case control
7
2148
5.0
22
0.16 (0.11 to 0.23)
6 (5 to 7)


The three types of study design had a similar result in that the relative risk was about 0.15, indicating a large reduction in UTI rates in circumcised over uncircumcised boys. There was consistency within the study types as well, and the four large cohort studies are shown in Figure 1. The absolute size of the effect varied, so that a number needed to treat calculated from the cohort studies was about 100, while that calculated from case-control studies was 6. The reason for the difference was the very large event rates in case control studies compared with cohorts; in uncircumcised boys the event rate was 20-fold higher in case control studies.



Figure 1: UTI rates in cohort studies, in circumcised and uncircumcised boys





A useful calculation was for benefit and harm for circumcision for UTI at different rates of urinary tract infection. Table 2 shows calculations in 1000 boys for three rates of UTI, a normal UTI rate of 1%, a higher rate of 10% in boys with recurrent UTI, and a highest rate of 30% in boys with vesicouretic reflux. Using the same 10-fold reduction with circumcision, and known information about complications at 2%, it is clear that the benefits of circumcision outweigh harm when UTI rates are higher than 2-3%.



Table 2: Benefit and harm of circumcision at different UTI rates



UTI numbers in 1000 patients
Number of
Patient group
Risk of UTI (%)
Circumcised
Uncircumcised
UTI prevented
Complications of circumcision
Normal
1
1
10
9
20
Previous UTI
10
13
100
87
20
Vesicouretic reflux
30
39
300
261
20


Comment

This is a useful and informative systematic review. Apparently there have been differences of opinion on how big the effect of circumcision is on UTI, though the individual studies here showed a remarkable consistency. Circumcision substantially reduces the rate of UTI, and when UTI rates are 3% or more benefits probably outweigh harms. And there is a lesson, that while the statistical results may be the same between study designs, the absolute rates can vary substantially, with real consequences for calculations of NNT.

Reference:

  1. D Singh-Grewal et al. Circumcision for the prevention of urinary tract infection in boys: a systematic review of randomised trials and observational studies. Archives of Diseases of Childhood 2005 90: 853-858.

previous or next story