Circumcision and Urinary Tract Infections

Circumcision and Urinary Tract Infection

Urinary Tract Infections (UTIs) are rare, and mainly occur in the first year of life. They are several times commoner in girls than boys (but of course surgery is, rightly, never considered for girls).

They are painful, and women's experience of them is a powerful inducement to have sons circumcised, if they imagine that this will protect them. In fact, a significant proportion of boys contract UTIs even though they are circumcised. A study in Israel found they mainly occured in girls at four months, but in boys soon after they were circumcised.... Now an Australian study suggests circumcision may cause urinary problems.

The supposed correlation between intactness and UTIs was only discovered in 1982, long after the custom of circumcision was well-established in the US. (Yet today it is commonly presented as the first reason, illustrating how circumcision is a "solution" looking for a problem - constantly looking, as each new problem is disposed of.)

Ironically, all the 1982 paper did was quite casually note that "95% of the [male] infants [with UTI] were uncircumcised." without mentioning that virtually no babies born at that hospital (Parkland in Dallas, Texas) were circumcised. The paper went on: "All infants responded promptly to antimicrobial therapy."

Dr Thomas Wiswell
It is Dr Thomas Wiswell who has made it his life's work to establish the UTI-intact penis connection, and he is also a keen circumcisor. His many papers present statistics involving huge samples, tens of thousands of babies (which may convince those who mistakenly think large numbers guarantee accuracy). However, all the babies were born in military hospitals, where circumcision is very much the norm. It is very likely that the intact babies had their foreskins forcibly retracted. It is also likely that a major reason for the babies to be left intact was that they were born prematurly, put in intensive care and catheterised for that reason: the catheterisation, not their intactness, caused their UTIs. There are also issues around the method of determining UTI infection. Some methods of collecting urine samples risk contamination. There is a more detailed rebuttal of Wiswell on the CIRP website.

The study evaluated charts of boys born at Army hospitals worldwide and looked at those who had urine cultures and whether they were circuncised or not. Wiswell, an Army neonatologist, decided that circumcised boys had fewer UTIs based on the greater number of positive cultures. In fact, the only variable was whether the boys were catheterised, since the circumcised boys' urine was obtained by clean catch bags, while the intact boys' urine was from catheterisation, thus growing whatever bacteria were hiding under the foreskin or on the skin of the glans itself - in other words, cultures were positive for harmless bacteria which were not causing infection, but were simply present, as they are on all of us. The cultures were irrelevant, but this started the whole claim that circumcision prevents UTI.

One commonly quoted figure is that UTIs are ten times as common in intact boys (1.1%) as circumcised boys (0.1%). This means that circumcision can offer an absolute reduction of 1% at best.

Combining the rarity of UTIs, with the only partial prevention attributable to circumcision, To et al calculated that it would take 195 circumcisions to prevent one UTI:

bar-chart: 1 UTI prevented vs 194 not prevented

This figure is conservative. One factor that might make it even higher is that doctors unfamiliar with the intact penis are inclined to forcibly retract the foreskin, and this may open the way to infection.

A 2007 paper is being quoted as supporting the claim that circumcision protects against UTIs, using absurdly shonky statistics. The circumcision status of only about half of the boys with UTIs - 36 boys, a tiny fraction of the total - is known, but that is used as if it proved something.

Another confounding factor is that hospitalisiation was taken as the defining degree of seriousness of UTIs: but babies with UTIs are often hospitalised in order to be circumcised, so of course more intact babies appear in the statistics.

The main factor correlating with UTIs in infants is urinary tract abnormalities. Breast-feeding is reported to reduce UTIs' incidence (because the close contact with the mother colonises the baby's body with her benign bacteria in preference to those that cause UTIs).

UTIs are readily treated with antibiotics.

Does circumcision cause disease?

Prevalence of urinary symptoms in urban Australian men aged 40-69

J Epidemiol Biostat 2001;6(2):211-8 (ISSN: 1359-5229)
McCredie M; Staples M; Johnson W; English DR; Giles GG
Department of Preventive and Social Medicine, Dunedin Medical School, University of Otago, New Zealand.

BACKGROUND: This study was devised to determine the prevalence of urinary symptoms among men living in the Australian cities of Melbourne, Sydney or Perth, and to identify factors associated with the presence of moderate-to-severe urinary symptoms.

METHODS: The study comprised a population-based sample of 1,216 men, aged 40-69 years, whose names were obtained through electoral rolls and who participated as controls in a case-control study of risk factors for prostate cancer. As part of a structured face-to-face interview, the men completed the International Prostate Symptom Score (IPSS). Men with moderate (IPSS = 8-19) or severe (IPSS > or = 20) urinary symptoms were compared with those with mild or no symptoms (IPSS < 8) using unconditional logistic regression.

RESULTS: The age-specific prevalence of moderate-to-severe urinary symptoms (IPSS > or = 8) in men aged 40-49, 50-59, 60-69 years was 16%, 23% and 28%, respectively. Compared with men with no or mild urinary symptoms (IPSS < 8), men with moderate-to-severe symptoms were more likely to report not currently living as married [odds ratio (OR) = 1.5; 95% confidence interval (CI) 1.1-2.0] and being circumcised (OR = 1.5; 95% Cl 1.2-2.0). The increased likelihood associated with drinking an average of > 60g per day of alcohol in the 2 years before interview was of marginal statistical significance (OR = 1.6; 1.0-2.6). There were no significant differences between men with IPSS > or = 8 and those with IPSS < 8 with respect to body mass index, education level, having had a vasectomy, or cigarette smoking.

CONCLUSION: Among Australian men, being circumcised, or not currently living as married, were associated with increased prevalence of urinary symptoms.

Back to the Intactivism index page

The first reference is by GM Ginsberg and GH McCracken, "Urinary tract infections in young children" Pediatrics 1982; 69:409-12, according to "Neonatal Male Circumcision", a policy statement by a working party of the Australian College of Paediatrics, August 1995.

The 2007 study is P.H.Conway, A. Cnaan, T. Zaoutis, B.V. Henry, R.W. Grundmeier, R. Keren, "Recurrent Urinary Tract Infections in Children, Risk Factors and Association With Prophylactic Antimicrobials," JAMA, July 11, 2007, Vol 298, No. 2

"From a network of 27 primary care pediatric practices in urban, suburban, and semirural areas spanning 3 states, a cohort of children aged 6 years or younger who were diagnosed with first UTI between July 1, 2001, and May 31, 2006, was assembled." ("States" implies the study was entirely US-based. The doctors' affiliations (Philadelphia and Cincinatti) suggest eastern states.)

	First UTI (n = 611)	Recurrent UTI (n = 83)
Sex
Male	68 (11.1)	8 (9.6)
Female	543 (88.9)	75 (90.4)

That's to say EIGHT times as many females got a first UTI as males and NINE times as many got a recurrent UTI (but the difference between first and recurrent UTIs is not significant: femaleness is not associated with recurrence as such. It's probably safe to assume that the sex ratio of the 74,974-strong cohort was about 50:50, but this is not specified. (The researchers also draw conclusions about race.)

The total information about circumcision is:

"Of the 68 male children [with any UTI], there was no documented circumcision status for 32 (47%). Twenty-six (38%) were uncircumcised and 10 (15%) were circumcised."

That is, they had no circumcision data for nearly half of those with a first UTI. This throws the data about the other half into meaninglessness, right there. There is also no data for the ~34,900 boys WITHOUT a first UTI, so those percentages tell us nothing. Suppose, for a ridiculous example, that of the 35,000 males only those 10 were circumcised, then the UTI rate among circumcised males was 10/10 or 100% and among the intact only 26/34,990 or 0.074%, and a good case could be made that circumcision causes UTIs and/or that intactness protects against them.

And later

"Among male children in whom circumcision status was known, 5 of 26 (19%) uncircumcised vs 0 of 10 circumcised children had a recurrent UTI (P=.13)." (P=.13 means no statistical significance.)

Without any information about the status of the whole sample, this is also meaningless. Taking a p value, or a percentage, for a sample of 36 children is nonsense. Common sense tells you that these figures are too small to show anything. Three more circumcised boys with recurrent UTIs (out of ~35,000, or even out of the 32 whose circumcision status was unknown) would be 23%, overturning any claim.

To print percentages of circumcised and intact with first and recurrent UTIs in the form Conway et al. did was misleading and irresponsible.

It is also quite likely that "lack of circumcision" was more likely to be noted when a boy had a UTI than when he didn't, because of the previous work of Wiswell et al. in claiming there is a link.

Yet Prof Brian Morris goes one step further. He has expressed this figure as

“Recurrent UTIs occur in 19% of uncircumcised boys, but in none of the circumcised.”

Morris, B, Why Circumcision is a biomedical imperative for the 21st Century, BioEssays 29:11, 1151,

(He divided (a) by (b) in the diagram above.) Here is a closeup, to scale.

It should be clear that "19%" is a perfectly nonsensical figure.

A statistician comments:

It is not possible to evaluate whether a treatment is worthwhile based on its Relative Risk Reduction. While a 90% reduction may sound impressive, 90% of a tiny amount is an even tinier amount. The Number Needed to Treat (NNT) is a much better indicator, because it makes it possible to compare the cost of the treatment with the benefit of the treatment.

The lower the incidence of a condition, the higher the NNT required to prevent that condition. Because the incidence of UTI is low, the NNT will be high irrespective of the relative risk reduction of the treatment. Even when the relative risk reduction is 90% the NNT is very high. The lower the relative risk reduction, the higher the NNT (and when the relative risk reduction is zero, even an infinite number of treatments will not prevent a UTI).

Even though we are not given either

the circumcision status of 32 of the 68 boys with a UTI, or
among the young males in the general population to which the subjects belonged

we can calculate NNTs for various values of these variables.

Of the 68 boys, we know that 26 were intact and 10 were circumcised, so the total number of intact boys must be between 26 and 58. If the circumcision statuses of an intact boy and a circumcised boy were equally likely to have been noted, then 49 of the 68 boys with a UTI were intact. However, where the circumcision rate is very high, it is plausible that intact status would be considered more noteworthy than circumcised status and thus more likely to have been noted, in which case the intact:circumcised ratio among the 68 boys with UTIs would be lower than the 26:10 ratio among the 36 boys whose statuses were noted.

It seems likely that the children in the study came from Ohio, Pennsylvania and one other (possibly adjoining) state in which the circumcision rates are very high.

A number of US studies have observed that UTI incidence among circumcised boys was lower than that among intact boys. They have failed to establish whether that is because circumcision reduced the incidence, or because improper care of the intact penis in the US elevates the incidence in intact boys (or both, or neither). Whatever the cause, conventional wisdom has it that the incidence is about 90% lower in circumcised boys.

Based on the data in the Conway study, it can be calculated that

if circumcision reduces UTI incidence by 90%, and
if the circumcision rate of young boys in the general population to which the study subjects belonged is in the range 78% to 87.5%,

then

it probably requires between 133 and 186 circumcisions to prevent a single UTI (almost always successfully treatable with antibiotics) and
it requires between 600 and 1,577 circumcisions to prevent a recurrent UTI.

It is possible that a large number of UTIs in intact boys in the US are caused by improper care of the intact penis, i.e., that the difference, if any, between the UTI incidence in intact boys and circumcised boys would be much smaller than has been observed in US studies if physicians and parents knew how to care for the intact penis. If we assume that 50% of the UTIs observed in intact males in the study were caused by improper care, and, if circumcision reduces UTI incidence by 80%, we can calculate that

it probably requires somewhere between 300 and 417 circumcisions to prevent a single UTI and
it requires between 1,200 and 3,548 circumcisions to prevent a recurrent UTI.

If this is the case, and if one costs a circumcision at $300 (and ignores the cost of re-circumcisions, treatment of complications and any consideration of long-term damage), it costs

of the order of $100,000 to prevent a single UTI (which can be treated by a ~$20 course of antibiotics), and
$300,000 to $1,000,000 to prevent a recurrent UTI.

The UTI incidence among young girls in the study was 1.48%, while the highest credible UTI incidence among intact young boys was 0.83% -- i.e., the rate anong girls was at least 78% higher than the rate among intact boys -- hardly a comparison upon which to base advocating prophylactic surgery on boys.